# hume.ai > AI-optimized mirror of hume.ai containing 276 pages totalling 164,949 words of clean markdown content, structured data, and semantic HTML. Original source: https://hume.ai. Last updated: 2026-07-19T02:46:16.655Z. Each page is available as HTML (with JSON-LD structured data) and Markdown (text-only, ideal for LLMs and RAG). ## Homepage - [Hume API Status](/content/status/index.html) (86 words) - [Chatter • Hume AI](/content/chatter/index.html): Chatter: An interactive podcast experience (20 words) - [Hume AI - Human Feedback for Voice, Speech, and Conversational AI | Hume AI](/content/site-root.html): Real human ratings, in a single API call. The human evaluation layer for voice, speech, and conversational AI. (475 words) ## Articles & Blog Posts - [api/v0/tts/file/index.html](/content/api/v0/tts/file/index.html) (4 words) - [Vercel AI SDK | Hume API](/content/dev/docs/integrations/vercel-ai-sdk/index.html): Guide to integrating Hume TTS into your web application with the Vercel AI SDK. (855 words) - [LiveKit | Hume API](/content/dev/docs/integrations/livekit/index.html): Guide to integrating Hume TTS via the Hume LiveKit Agents plugin. (828 words) - [Vapi | Hume API](/content/dev/docs/integrations/vapi/index.html): Use Hume as a text-to-speech (TTS) voice provider in your Vapi assistant. (922 words) - [Twilio | Hume API](/content/dev/docs/integrations/twilio/index.html): Guide to enabling phone calling with the Empathic Voice Interface (EVI). (946 words) - [Agora | Hume API](/content/dev/docs/integrations/agora/index.html): Guide to integrating Hume TTS with Agora's Conversational AI Engine. (573 words) - [Voice Collections](/content/app/voices/collections/featured/index.html): The leading developer platform for expressive communication (337 words) - [Page Not Found | Hume AI](/content/policies/acceptable-use-policy/index.html): The page you're looking for doesn't exist. (21 words) - [Hume AI • Voice Library](/content/app/voices/index.html): The leading developer platform for expressive communication (464 words) - [Human Feedback API - Hume AI | Hume AI](/content/human-feedback-api/index.html): Launch studies, collect real human feedback, and get results in hours. The human evaluation API for voice, speech, and conversational AI. (457 words) - [Expression Measurement API - Hume AI | Hume AI](/content/expression-measurement-api/index.html): Identify emotion in voice, offline or in real time. The Tagger API returns 600+ expression dimensions from any audio; the Prosody API delivers real-time emotional signals during live conversations. (380 words) - [Datasets - Voice & Expression Data Built by Researchers | Hume AI | Hume AI](/content/datasets/index.html): Extensive, curated, and scientifically validated voice and expression datasets designed to solve the hardest problems in speech, emotion, and multimodal AI. (307 words) - [Kairos - Hume AI | Hume AI](/content/kairos/index.html): Simulate, evaluate, and benchmark your voice AI the way humans experience it - real-world scenario simulation, automated evals, and human-grounded studies. Bring any model, get results in hours. (351 words) - [Contact Our Research Team - Hume AI | Hume AI](/content/sales-form/index.html): Contact our research team for licensable training datasets, expert guidance, and partnerships to build expressive voice AI. (124 words) - [Hume AI • Sign Up](/content/app/sign-up/index.html): The leading developer platform for expressive communication (32 words) - [EVI .NET Quickstart | Hume API](/content/dev/docs/speech-to-speech-evi/quickstart/dotnet/index.html): A quickstart guide for integrating the Empathic Voice Interface (EVI) with .NET. (1,038 words) - [Tool Use | Hume API](/content/dev/docs/speech-to-speech-evi/features/tool-use/index.html): Guide to utilize function calling with EVI. (2,580 words) - [EVI Version | Hume API](/content/dev/docs/speech-to-speech-evi/configuration/evi-version/index.html): How to set and update the EVI version in your configuration. (746 words) - [Changelog | Hume API](/content/dev/changelog/index.html) (3,465 words) - [Audio | Hume API](/content/dev/docs/speech-to-speech-evi/guides/audio/index.html): Guide to recording and playing audio for an EVI chat. (2,280 words) - [Webhooks | Hume API](/content/dev/docs/speech-to-speech-evi/configuration/webhooks/index.html): Guide to EVI's webhooks configuration option. (1,468 words) - [Chat History | Hume API](/content/dev/docs/speech-to-speech-evi/features/chat-history/index.html): Guide to accessing chat history with the EVI. (1,211 words) - [Custom Language Model | Hume API](/content/dev/docs/speech-to-speech-evi/guides/custom-language-model/index.html): Use a custom language model to generate your own text, for maximum configurability. (1,774 words) - [Chat | Hume API](/content/dev/reference/speech-to-speech-evi/chat/index.html): Chat with Empathic Voice Interface (EVI) (853 words) - [Create prompt | Hume API](/content/dev/reference/speech-to-speech-evi/prompts/create-prompt/index.html): Creates a Prompt that can be added to an EVI configuration(/reference/speech-to-speech-evi/configs/create-config). (330 words) - [Create config version | Hume API](/content/dev/reference/speech-to-speech-evi/configs/create-config-version/index.html): Updates a Config by creating a new version of the Config. (525 words) - [List chat events | Hume API](/content/dev/reference/speech-to-speech-evi/chats/list-chat-events/index.html): Fetches a paginated list of Chat events. (278 words) - [Send Message | Hume API](/content/dev/reference/speech-to-speech-evi/control-plane/send/index.html): Send a message to a specific chat. (426 words) - [Get chat group audio | Hume API](/content/dev/reference/speech-to-speech-evi/chat-groups/get-audio/index.html): Fetches a paginated list of audio for each Chat within the specified Chat Group. (294 words) - [Get chat_group | Hume API](/content/dev/reference/speech-to-speech-evi/chat-groups/get-chat-group/index.html): Fetches a ChatGroup by ID, including a paginated list of Chats associated with the ChatGroup. (345 words) - [Create config | Hume API](/content/dev/reference/speech-to-speech-evi/configs/create-config/index.html): Creates a Config which can be applied to EVI. (645 words) - [TTS NodeJS Quickstart Guide | Hume API](/content/dev/docs/text-to-speech-tts/quickstart/typescript/index.html): Step-by-step guide for integrating the TTS API using Hume's TypeScript SDK. (1,613 words) - [List tool versions | Hume API](/content/dev/reference/speech-to-speech-evi/tools/list-tool-versions/index.html): Fetches a list of a Tool's versions. (238 words) - [Create tool | Hume API](/content/dev/reference/speech-to-speech-evi/tools/create-tool/index.html): Creates a Tool that can be added to an EVI configuration(/reference/speech-to-speech-evi/configs/create-config). (462 words) - [Update tool description | Hume API](/content/dev/reference/speech-to-speech-evi/tools/update-tool-description/index.html): Updates the description of a specified Tool version. (329 words) - [List prompt versions | Hume API](/content/dev/reference/speech-to-speech-evi/prompts/list-prompt-versions/index.html): Fetches a list of a Prompt's versions. See our prompting guide(/docs/speech-to-speech-evi/guides/phone-calling) for tips on crafting your system prompt. (224 words) - [Get config version | Hume API](/content/dev/reference/speech-to-speech-evi/configs/get-config-version/index.html): Fetches a specified version of a Config. (406 words) - [Update config description | Hume API](/content/dev/reference/speech-to-speech-evi/configs/update-config-description/index.html): Updates the description of a Config. (442 words) - [Create prompt version | Hume API](/content/dev/reference/speech-to-speech-evi/prompts/create-prompt-version/index.html): Updates a Prompt by creating a new version of the Prompt. (335 words) - [Chat Started | Hume API](/content/dev/reference/speech-to-speech-evi/chat-webhooks/chat-started/index.html): Sent when an EVI chat is started. (177 words) - [List chat events from a specific chat_group | Hume API](/content/dev/reference/speech-to-speech-evi/chat-groups/list-chat-group-events/index.html): Fetches a paginated list of Chat events associated with a Chat Group. (239 words) - [List chats | Hume API](/content/dev/reference/speech-to-speech-evi/chats/list-chats/index.html): Fetches a paginated list of Chats. (260 words) - [Tool Call | Hume API](/content/dev/reference/speech-to-speech-evi/chat-webhooks/tool-call/index.html): Sent when EVI triggers a tool call (175 words) - [List chat_groups | Hume API](/content/dev/reference/speech-to-speech-evi/chat-groups/list-chat-groups/index.html): Fetches a paginated list of Chat Groups. (270 words) - [Chat Ended | Hume API](/content/dev/reference/speech-to-speech-evi/chat-webhooks/chat-ended/index.html): Sent when an EVI chat ends. (219 words) - [Update prompt description | Hume API](/content/dev/reference/speech-to-speech-evi/prompts/update-prompt-description/index.html): Updates the description of a Prompt. See our prompting guide(/docs/speech-to-speech-evi/guides/phone-calling) for tips on crafting your system prompt. (280 words) - [Create tool version | Hume API](/content/dev/reference/speech-to-speech-evi/tools/create-tool-version/index.html): Updates a Tool by creating a new version of the Tool. (340 words) - [Get tool version | Hume API](/content/dev/reference/speech-to-speech-evi/tools/get-tool-version/index.html): Fetches a specified version of a Tool. (352 words) - [Get prompt version | Hume API](/content/dev/reference/speech-to-speech-evi/prompts/get-prompt-version/index.html): Fetches a specified version of a Prompt. See our prompting guide(/docs/speech-to-speech-evi/guides/phone-calling) for tips on crafting your system prompt. (292 words) - [Stream Input | Hume API](/content/dev/reference/text-to-speech-tts/stream-input/index.html): Generate emotionally expressive speech. (356 words) - [Get chat audio | Hume API](/content/dev/reference/speech-to-speech-evi/chats/get-audio/index.html): Fetches the audio of a previous Chat. (142 words) - [List voices | Hume API](/content/dev/reference/voices/list/index.html): Lists voices you have saved in your account, or voices from the Voice Library(https://app.hume.ai/tts/voice-library). (259 words) - [List prompts | Hume API](/content/dev/reference/speech-to-speech-evi/prompts/list-prompts/index.html): Fetches a paginated list of Prompts. See our prompting guide(/docs/speech-to-speech-evi/guides/phone-calling) for tips on crafting your system prompt. (240 words) - [List configs | Hume API](/content/dev/reference/speech-to-speech-evi/configs/list-configs/index.html): Fetches a paginated list of Configs. (244 words) - [Update config name | Hume API](/content/dev/reference/speech-to-speech-evi/configs/update-config-name/index.html): Updates the name of a Config. (60 words) - [Errors | Hume API](/content/dev/docs/resources/errors/index.html) (1,993 words) - [List tools | Hume API](/content/dev/reference/speech-to-speech-evi/tools/list-tools/index.html): Fetches a paginated list of Tools. (147 words) - [Update tool name | Hume API](/content/dev/reference/speech-to-speech-evi/tools/update-tool-name/index.html): Updates the name of a Tool. (97 words) - [Delete tool | Hume API](/content/dev/reference/speech-to-speech-evi/tools/delete-tool/index.html): Deletes a Tool and its versions. (48 words) - [Delete tool version | Hume API](/content/dev/reference/speech-to-speech-evi/tools/delete-tool-version/index.html): Deletes a specified version of a Tool. (102 words) - [List config versions | Hume API](/content/dev/reference/speech-to-speech-evi/configs/list-config-versions/index.html): Fetches a list of a Config's versions. (257 words) - [Delete prompt version | Hume API](/content/dev/reference/speech-to-speech-evi/prompts/delete-prompt-version/index.html): Deletes a specified version of a Prompt. See our prompting guide(/docs/speech-to-speech-evi/guides/phone-calling) for tips on crafting your system prompt. (97 words) - [Delete prompt | Hume API](/content/dev/reference/speech-to-speech-evi/prompts/delete-prompt/index.html): Deletes a Prompt and its versions. See our prompting guide(/docs/speech-to-speech-evi/guides/phone-calling) for tips on crafting your system prompt. (38 words) - [Update prompt name | Hume API](/content/dev/reference/speech-to-speech-evi/prompts/update-prompt-name/index.html): Updates the name of a Prompt. See our prompting guide(/docs/speech-to-speech-evi/guides/phone-calling) for tips on crafting your system prompt. (56 words) - [Voice Conversion (Streamed JSON) | Hume API](/content/dev/reference/text-to-speech-tts/convert-voice-json/index.html) (194 words) - [Delete config version | Hume API](/content/dev/reference/speech-to-speech-evi/configs/delete-config-version/index.html): Deletes a specified version of a Config. (97 words) - [Voice Conversion (Streamed File) | Hume API](/content/dev/reference/text-to-speech-tts/convert-voice-file/index.html) (164 words) - [Delete config | Hume API](/content/dev/reference/speech-to-speech-evi/configs/delete-config/index.html): Deletes a Config and its versions. (44 words) - [Control Plane | Hume API](/content/dev/reference/speech-to-speech-evi/control-plane/chat-chat-id-connect/index.html): Connects to an in-progress EVI chat session. The original chat must have been started with allowconnection=true. (84 words) - [Delete voice | Hume API](/content/dev/reference/voices/delete/index.html): Deletes a previously generated custom voice. (25 words) - [TTS .NET Quickstart Guide | Hume API](/content/dev/docs/text-to-speech-tts/quickstart/dotnet/index.html): Step-by-step guide for integrating the TTS API using Hume's .NET SDK. (1,132 words) - [Audio Reconstruction | Hume API](/content/dev/docs/speech-to-speech-evi/features/audio-reconstruction/index.html): Guide to reconstructing the audio from previous Chats for playback. (521 words) - [TTS Python Quickstart Guide | Hume API](/content/dev/docs/text-to-speech-tts/quickstart/python/index.html): Step-by-step guide for integrating the TTS API using Hume's Python SDK. (1,007 words) - [Hume MCP Server | Hume API](/content/dev/docs/integrations/mcp/index.html): Use Hume AI's Octave TTS with your favorite MCP clients like Claude Desktop, Cursor, and Windsurf. (1,240 words) - [Text-to-Speech (Streamed JSON) | Hume API](/content/dev/reference/text-to-speech-tts/synthesize-json-streaming/index.html): Streams synthesized speech using the specified voice. If no voice is provided, a novel voice will be generated dynamically. (598 words) - [Text-to-Speech (Streamed File) | Hume API](/content/dev/reference/text-to-speech-tts/synthesize-file-streaming/index.html): Streams synthesized speech using the specified voice. If no voice is provided, a novel voice will be generated dynamically. (541 words) - [EVI TypeScript Quickstart | Hume API](/content/dev/docs/speech-to-speech-evi/quickstart/typescript/index.html): A quickstart guide for integrating the Empathic Voice Interface (EVI) with TypeScript. (761 words) - [Continuation Guide | Hume API](/content/dev/docs/text-to-speech-tts/continuation/index.html): Guide to maintaining coherent speech across multiple utterances and generations. (887 words) - [Control Plane | Hume API](/content/dev/docs/speech-to-speech-evi/guides/control-plane/index.html): Guide to controlling active EVI chats from a trusted backend, including updating session settings and streaming mirrored events. (547 words) - [EVI Next.js Quickstart | Hume API](/content/dev/docs/speech-to-speech-evi/quickstart/nextjs/index.html): A quickstart guide for implementing the Empathic Voice Interface (EVI) with Next.js. (709 words) - [Voice Design | Hume API](/content/dev/docs/voice/voice-design/index.html): A guide to designing expressive, natural-sounding voices using Octave, Hume's speech-language model. (1,164 words) - [Empathic Voice Interface FAQ | Hume API](/content/dev/docs/speech-to-speech-evi/faq/index.html) (1,263 words) - [EVI Python Quickstart | Hume API](/content/dev/docs/speech-to-speech-evi/quickstart/python/index.html): A quickstart guide for integrating the Empathic Voice Interface (EVI) with Python. (699 words) - [Prompt Engineering for EVI | Hume API](/content/dev/docs/speech-to-speech-evi/guides/prompting/index.html): Guide to crafting system prompts to shape the behavior, responses, and style of the Empathic Voice Interface (EVI). (559 words) - [Session Settings | Hume API](/content/dev/docs/speech-to-speech-evi/configuration/session-settings/index.html): Guide to applying EVI configuration options in real time with Session Settings. (739 words) - [Pipecat | Hume API](/content/dev/docs/integrations/pipecat/index.html): Guide to integrating Hume TTS via the Pipecat framework. (720 words) - [Text-to-Speech (File) | Hume API](/content/dev/reference/text-to-speech-tts/synthesize-file/index.html): Synthesizes one or more input texts into speech using the specified voice. If no voice is provided, a novel voice will be generated dynamically. (261 words) - [Voice Cloning | Hume API](/content/dev/docs/voice/voice-cloning/index.html): Create custom voice clones from speech using Octave, either by recording your voice or uploading a sample. (500 words) - [Pause Responses | Hume API](/content/dev/docs/speech-to-speech-evi/features/pause-responses/index.html): Guide to pausing EVI's responses during a Chat session. (404 words) - [Voice | Hume API](/content/dev/docs/speech-to-speech-evi/configuration/voice/index.html): Guide to configuring the voice of the Empathic Voice Interface (EVI). (445 words) - [Resuming Chats | Hume API](/content/dev/docs/speech-to-speech-evi/features/resume-chats/index.html): Guide to preserving context from previous Chat sessions. (200 words) - [Turn detection | Hume API](/content/dev/docs/speech-to-speech-evi/configuration/turn-detection/index.html): Tune voice activity detection and end-of-turn behavior. (384 words) - [TTS CLI Quickstart Guide | Hume API](/content/dev/docs/text-to-speech-tts/quickstart/cli/index.html): Step-by-step guide for integrating the TTS API using Hume's CLI. (550 words) - [Acting Instructions Guide | Hume API](/content/dev/docs/text-to-speech-tts/acting-instructions/index.html): Guide to controlling voice expression in Octave TTS through acting instructions, speed settings, and silence parameters. (653 words) - [Privacy | Hume API](/content/dev/docs/resources/privacy/index.html) (725 words) - [Dynamic Variables | Hume API](/content/dev/docs/speech-to-speech-evi/features/dynamic-variables/index.html): Personalize EVI chats by leveraging dynamic variables in your system prompt. (206 words) - [Interruptibility | Hume API](/content/dev/docs/speech-to-speech-evi/features/interruptibility/index.html): Guide to EVI's interruptibility feature and how to manage interruptions on the client. (183 words) - [Voice Management | Hume API](/content/dev/docs/voice/management/index.html): A guide to viewing, renaming, and deleting custom voices via the Platform UI or API. (288 words) - [Timeouts | Hume API](/content/dev/docs/speech-to-speech-evi/configuration/timeouts/index.html): Define and manage chat session duration limits. (195 words) - [Language Model | Hume API](/content/dev/docs/speech-to-speech-evi/configuration/language-model/index.html): Choose which language model to use for EVI's response generation. (319 words) - [Interruption | Hume API](/content/dev/docs/speech-to-speech-evi/configuration/interruption/index.html): Configure EVI's interruption sensitivity. (232 words) - [Timestamps Guide | Hume API](/content/dev/docs/text-to-speech-tts/timestamps/index.html): Guide to leveraging timestamps for audio outputted by Octave 2. (251 words) - [Tools | Hume API](/content/dev/docs/speech-to-speech-evi/configuration/tools/index.html): Equip EVI with Tools to enable function calling during Chats. (268 words) - [Event Messages | Hume API](/content/dev/docs/speech-to-speech-evi/configuration/event-messages/index.html): Schedule and author EVI responses for key Chat events. (253 words) - [Billing | Hume API](/content/dev/docs/resources/billing/index.html) (309 words) - [Voice Conversion | Hume API](/content/dev/docs/text-to-speech-tts/voice-conversion/index.html): Guide to converting your audio recordings into different voices using Octave's voice conversion API. (481 words) - [Text-to-Speech API FAQ | Hume API](/content/dev/docs/text-to-speech-tts/faq/index.html): We’ve compiled a list of frequently asked questions from our developer community. If you don’t see your question here, join the discussion on our Discord. (537 words) - [Use case guidelines | Hume API](/content/dev/docs/resources/use-case-guidelines/index.html) (445 words) - [Getting your API keys | Hume API](/content/dev/docs/introduction/api-key/index.html): Learn how to obtain your API keys and understand the supported authentication strategies for securely accessing Hume APIs. (483 words) - [Voice Guide | Hume API](/content/dev/docs/text-to-speech-tts/voice/index.html): Guide to using a saved voice or a Voice Library voice in your TTS API requests. (286 words) - [Voice | Hume API](/content/dev/docs/voice/overview/index.html): Utilize Hume’s Voice Library or design custom voices tailored to your application. (432 words) - [Model Architecture — Access the Architectures Behind Leading Voice Models | Hume AI | Hume AI](/content/model-architecture/index.html): Leapfrog years of R&D. Access TADA, EVI, and Octave — the architectures behind leading voice models. (824 words) - [Support | Hume API](/content/dev/support/index.html): Get technical support, contact our team, or explore enterprise and research programs. (239 words) - [Create voice | Hume API](/content/dev/reference/voices/create/index.html): Saves a new custom voice to your account using the specified TTS generation ID. (206 words) - [Welcome to Hume AI | Hume API](/content/dev/intro/index.html): Hume AI builds AI models that enable technology to communicate with empathy and support human well-being. (507 words) - [Training Data - Voice & Expression Datasets Built by Researchers | Hume AI | Hume AI](/content/training-data/index.html): Extensive, curated, and scientifically validated voice and expression datasets designed to solve the hardest problems in speech, emotion, and multimodal AI. (335 words) - [Speech-to-Speech (EVI) | Hume API](/content/dev/docs/speech-to-speech-evi/overview/index.html): Hume's Empathic Voice Interface (EVI) is an advanced, real-time emotionally intelligent voice AI. (780 words) - [Hume AI • Reset Password](/content/app/reset-password/index.html): The leading developer platform for expressive communication (24 words) - [Context Injection | Hume API](/content/dev/docs/speech-to-speech-evi/features/context-injection/index.html): Guide to injecting context during an EVI Chat session. (345 words) - [Configuring EVI | Hume API](/content/dev/docs/speech-to-speech-evi/configuration/build-a-configuration/index.html): Guide to configuring the Empathic Voice Interface (EVI). (1,370 words) - [Text-to-Speech (Json) | Hume API](/content/dev/reference/text-to-speech-tts/synthesize-json/index.html): Synthesizes one or more input texts into speech using the specified voice. If no voice is provided, a novel voice will be generated dynamically. (538 words) - [System Prompt | Hume API](/content/dev/docs/speech-to-speech-evi/configuration/system-prompt/index.html): Define EVI’s behavior, responses, and style. (203 words) - [Text-to-Speech (TTS) | Hume API](/content/dev/docs/text-to-speech-tts/overview/index.html): Introduction to Hume's TTS API, including its features, usage limits, and key concepts for integration. (971 words) - [Octave - Text-to-Speech with Emotional Intelligence | Hume AI](/content/octave/index.html): Generate expressive, natural-sounding speech that conveys the full range of human emotion. Text-to-speech powered by emotional AI. (841 words) - [Terms of Use | Hume AI | Hume AI](/content/terms-of-use/index.html): Terms of Use for Hume AI Platform, APIs, and services. Read our terms and conditions for using our empathic AI technology. (1,211 words) - [Creator Studio - Transform Documents into Production-Ready Audio | Hume AI](/content/creator-studio/index.html): Import your content, assign voices, add acting directions, and export production-ready audio. Create hours of natural-sounding speech for audiobooks, videos, and more. (707 words) - [Customers - Companies Using Hume AI for Empathic Experiences | Hume AI](/content/customers/index.html): See how leading companies are using Hume to build empathic AI experiences. Discover customer success stories and use cases. (232 words) - [Blog - Insights on Empathic AI and Emotional Intelligence | Hume AI](/content/blog/index.html): Explore articles, case studies, product updates, and research from Hume AI. Stay updated on the latest developments in empathic AI and emotional intelligence. (259 words) - [Privacy Policy | Hume AI | Hume AI](/content/privacy-policy/index.html): Privacy Policy for Hume AI. Learn how we collect, use, and protect your personal information when using our empathic AI services. (634 words) - [SLM Judge Leaderboard - Hume AI | Hume AI](/content/slm-judge-leaderboard/index.html): A leaderboard of the judges themselves: how closely each SLM judge's scores track human ratings on expressive text-to-speech evals, measured as Spearman agreement. (556 words) - [Developers - Build with Emotional Intelligence | Hume AI | Hume AI](/content/developers/index.html): Everything you need to add emotional AI to your applications. SDKs for every platform, integrations with your favorite tools, and documentation to get you shipping fast. (402 words) - [Explore AI Models | Hume AI | Hume AI](/content/explore/speech-prosody-model/index.html): Explore Hume AI's expression measurement models including facial expression analysis, speech prosody, and vocal expression detection. (134 words) - [Conversational AI - Build Empathic Voice and Chat Agents | Hume AI](/content/conversational-ai/index.html): Build AI agents that engage in natural, emotionally intelligent conversations. Create voice and chat experiences that understand context, adapt to user sentiment, and respond with genuine empathy. (392 words) - [Explore AI Models | Hume AI | Hume AI](/content/explore/facial-expression-model/index.html): Explore Hume AI's expression measurement models including facial expression analysis, speech prosody, and vocal expression detection. (127 words) - [Explore AI Models | Hume AI | Hume AI](/content/explore/vocal-expression-description-model/index.html): Explore Hume AI's expression measurement models including facial expression analysis, speech prosody, and vocal expression detection. (150 words) - [Explore AI Models | Hume AI | Hume AI](/content/explore/facs/index.html): Explore Hume AI's expression measurement models including facial expression analysis, speech prosody, and vocal expression detection. (160 words) - [Explore AI Models | Hume AI | Hume AI](/content/explore/vocal-expression-model/index.html): Explore Hume AI's expression measurement models including facial expression analysis, speech prosody, and vocal expression detection. (126 words) - [Acceptable Use Policy | Hume AI | Hume AI](/content/acceptable-use-policy/index.html): Guidelines for using Hume AI's services in a safe, lawful, and respectful manner. (307 words) - [Explore AI Models | Hume AI | Hume AI](/content/explore/dynamic-reaction/index.html): Explore Hume AI's expression measurement models including facial expression analysis, speech prosody, and vocal expression detection. (123 words) - [API Data Usage Policy | Hume AI | Hume AI](/content/api-data-usage-policy/index.html): Learn how Hume AI handles data submitted through our APIs, including our commitment to not using customer data for training our models. (183 words) - [Subprocessors | Hume AI | Hume AI](/content/subprocessors/index.html): List of subprocessors that Hume AI uses to carry out processing activities on customer data. (102 words) - [Publications - Research Papers from Hume AI | Hume AI](/content/publications/index.html): Research papers and academic publications from the Hume team. Explore our scientific contributions to empathic AI and emotional intelligence research. (2,145 words) - [Empathic Voice Interface (EVI) - AI Conversations with Emotional Intelligence | Hume AI](/content/empathic-voice-interface/index.html): Build voice-first AI experiences that understand and respond to human emotions. EVI enables natural, empathic conversations at scale. (785 words) - [Speech Data - Hume AI | Hume AI](/content/speech-data/index.html): We research and optimize audio datasets for other frontier voice AI labs. Scale your audio pre-training and post-training for speech-language models. (969 words) - [Real World VoiceEQ Bench - Hume AI | Hume AI](/content/rw-voice-eq/index.html): Voice AI, evaluated across every dimension that matters: speech recognition, speech understanding, text-to-speech quality, and live voice agent performance, scored on the conditions that decide real-world deployment. (690 words) - [Careers at Hume AI - Build the Future of Empathic AI | Hume AI](/content/careers/index.html): Join us in building AI that understands and responds to human emotions with care and nuance. We're on a mission to ensure artificial intelligence is developed in ways that are beneficial to humanity. (303 words) - [Pricing | Hume AI](/content/pricing/index.html): Explore our flexible pricing plans designed to fit your needs. Compare features, choose the best option, and get started today. (286 words) - [research/index.html](/content/research/index.html) (17 words) - [Introducing Real World VoiceEQ: Measuring the Human Quality of Voice AI | Hume Blog | Hume AI](/content/blog/introducing-real-world-voiceeq-measuring-the-human-quality-of-voice-ai.html) (1,053 words, Jul 14, 2026) - [Emotional Intelligence Is a Training-Time Property, Not a Prompt | Hume Blog | Hume AI](/content/blog/emotional-intelligence-is-a-training-time-property-not-a-prompt.html): You can't prompt your way to emotional intelligence — perception and expression live in the weights, not the system prompt. The scarce asset is the human-grounded reward signal that trains them in. (885 words, Jul 1, 2026) - [Speech-to-Speech Is the Hardest Problem in Voice AI | Hume Blog | Hume AI](/content/blog/speech-to-speech-is-the-hardest-problem-in-voice-ai.html): Speech-to-speech models promise to break the ceiling of the transcript-based voice pipeline. The architecture is right — but promise isn't production, and demos aren't evidence. Measurement is the deciding input. (879 words, Jun 1, 2026) - [Disentangling Emotion from Voice: A Cross-Product Sampling Approach for Expressive Voice Data | Hume Blog | Hume AI](/content/blog/disentangling-emotion-from-voice/index.html): Humans can separate how they feel from how they speak. Voice models still struggle to do the same. (1,343 words, May 27, 2026) - [The Science of What a Voice Reveals | Hume Blog | Hume AI](/content/blog/the-science-of-what-a-voice-reveals/index.html): Emotion isn't six categories — it's a continuous, high-dimensional space, carried as much by the voice as the words. Measuring it correctly is the foundation voice AI is built on. (816 words, May 4, 2026) - [Voice Models Are Commoditizing. The Value Is Moving Up the Stack. | Hume Blog | Hume AI](/content/blog/voice-models-are-commoditizing-the-value-is-moving-up-the-stack.html): As voice models converge and commoditize, model quality stops being the moat. Value migrates to the layer that measures, rigorously and continuously, how human a voice AI actually is. (777 words, Apr 15, 2026) - [Introducing the ACII 2026 Dyadic Contest (DaiKon) Workshop & Challenge | Hume Blog | Hume AI](/content/blog/acii-2026-daikon-challenge/index.html) (421 words, Apr 9, 2026) - [Opensourcing TADA: Fast, Reliable Speech Generation Through Text-Acoustic Synchronization | Hume Blog | Hume AI](/content/blog/opensource-tada/index.html): TADA (Text-Acoustic Dual Alignment) is Hume AI's open-source speech-language model that synchronizes text and audio one-to-one. (924 words, Mar 10, 2026) - [Voice Is Becoming AI's Primary Interface. Our Evaluation Methods Haven't Caught Up. | Hume Blog | Hume AI](/content/blog/voice-is-becoming-ai-s-primary-interface-our-evaluation-methods-haven-t-caught-up.html): Voice AI can transcribe accurately and respond fast, but conversations still feel brittle. The problem isn't the words — it's that our evaluation methods were never built to measure them. (831 words, Feb 9, 2026) - [Building Voice Models Is No Longer a Modeling Problem | Hume Blog | Hume AI](/content/blog/data-blog-jan/index.html): What’s changed isn’t just where voice is used, but what it represents. Voice is no longer a feature layered on top of an intelligent system. It’s becoming a foundational modality through which models reason, interact, and are judged by users. (941 words, Jan 21, 2026) - [Transforming AI Phone Support with WebAppClouds | Hume Blog | Hume AI](/content/blog/case-study-hume-webappclouds/index.html): EVI's conversational naturalness makes it exceptional for phone support. It recognizes tone, emotion, and phrasing, keeping interactions warm and efficient. The platform's intelligent end-of-turn detection ensures conversations flow smoothly without awkward pauses or interruptions. For WebAppClouds' clients, this translates to support customers actually enjoy. Instead of navigating complicated menus, customers simply say what they need, whether it's booking, rescheduling, or getting information. EVI integrates seamlessly with the WebAppClouds system, enabling real-time scheduling updates while delivering faster responses and consistent brand experiences. (308 words, Nov 6, 2025) - [GAF Powers Professional Training with Hume’s Text-to-Speech | Hume Blog | Hume AI](/content/blog/case-study-hume-gaf/index.html): To support their extensive training programs and marketing initiatives, GAF leverages Hume's text-to-speech technology to make internal training videos and marketing voiceovers. Our partnership addresses several key needs: Professional training content: Delivering consistent, high-quality audio for thousands of contractors and employees. Marketing collateral: Producing engaging voiceovers for promotional content and product demonstrations. Scalable production: Generating content without the logistics and cost of traditional voice recording. Hume's voice design also proved ideal for GAF. The platform's natural, expressive voices maintain the authoritative yet approachable tone that GAF needs to communicate with contractors, retailers, and customers. Unlike synthetic voices that can sound robotic or overly casual, Hume's TTS technology delivers the polished, trustworthy quality expected from an industry leader. (225 words, Nov 3, 2025) - [AudioStack x Hume: Professional Audio for Creatives | Hume Blog | Hume AI](/content/blog/case-study-hume-audiostack/index.html): AudioStack offers a comprehensive voice library for audio advertisements, podcasts, and branded content. As they continue to grow their voice offerings, they're integrating Hume's emotionally intelligent voices to meet two core demands of creative teams: 1. Consistent Stability Enterprise content generation requires voices that perform reliably across thousands of productions. Hume's voices deliver consistent quality and pronunciation, ensuring brand messaging remains clear and professional, whether creating one ad or thousands of dynamic variations. 2. Natural Expressiveness Generic TTS voices often sound flat or robotic—a dealbreaker for agencies creating audio that needs to engage audiences. Hume's voices bring genuine emotional depth, helping audio content feel authentic, engaging, and human. (233 words, Oct 24, 2025) - [Journee x Hume: Giving Enterprise AI Agents a Voice | Hume Blog | Hume AI](/content/blog/case-study-hume-journee/index.html): Hume replaced OpenAI, Deepgram, ElevenLabs, and Langfuse, more than halving Journee’s costs and dramatically simplifying debugging and prompt management. EVI solved Journee's challenges across five dimensions: Affordable Scaling: Industry-low pricing enabled Journee to deploy voice agents across every customer touchpoint. Clean APIs: Hume’s complete, well-documented APIs and Python SDK reduced integration time from weeks to days. Consistent Enterprise Performance: EVI delivers latency ranging from 140 ms to 1.3 s, outperforming any other solution Journee tested, even under heavy multi-session load. Technical Precision: Hume’s API sends a reliable “end-of-speech” signal, allowing Journee to instantly release inference resources. Expressive and Natural Voices: EVI sends the first second of speech almost instantly, then streams the rest reliably, enabling natural turn-taking and emotional realism. (374 words, Oct 23, 2025) - [Creating immersive avatar experiences with Render Foundry | Hume Blog | Hume AI](/content/blog/case-study-hume-render-foundry/index.html): Using Hume's custom voice cloning technology, Render Foundry created a Babe Ruth simulator that feels emotionally authentic. Hume's captured the tonal qualities, cadence, and personality of the baseball icon, allowing visitors to have natural, engaging conversations with one of sports' most beloved figures. The result is an experience that transcends typical museum exhibits. Visitors not only learn about Babe Ruth, but also connect with him. Render Foundry created something truly special by simulating Babe’s likeness and syncing the audio and the visual, making this experience one-of-a-kind. (253 words, Oct 21, 2025) - [Niantic Spatial x Hume AI: Creating Interactive & Spatially Aware AI Companions | Hume Blog | Hume AI](/content/blog/case-study-hume-niantic/index.html): In partnership with Snap Inc. (hardware) and Hume AI (voice), Niantic Spatial has developed location-aware companions for Spectacles, blending Snap Inc.’s AR glasses, Niantic Spatial’s Large Geospatial Model, and Hume’s Empathic Voice Interface (EVI) for natural, emotionally intelligent conversation. Niantic Spatial, the team pioneering AI that understands the physical world, is showcasing a compelling glimpse of what can happen when spatial intelligence and augmented reality meet. (519 words, Oct 16, 2025) - [Revelum x Hume: Detecting Voice Fraud in Real-Time | Hume Blog | Hume AI](/content/blog/case-study-hume-revelum/index.html): Our collaboration with Revelum creates a critical feedback loop for responsible AI development: Early Access to EVI: Revelum trains their models on Hume's cutting-edge voice technology, ensuring detection capabilities stay ahead of emerging threats before they reach malicious actors. Continuous Refinement: As Hume's emotionally intelligent voices become more sophisticated, Revelum's detection algorithms evolve in parallel. “By partnering with Hume, we’re taking a vital step toward building technology that anticipates — not just reacts to — the evolving tactics of bad actors seeking to misuse powerful models. Together, we’re staying one step ahead in ensuring generative AI is used responsibly.” (293 words, Oct 14, 2025) - [Creating Podcasts at Scale with Inception Point | Hume Blog | Hume AI](/content/blog/case-study-hume-inception-point/index.html): As Inception Point scales to thousands of AI personalities, they chose Hume’s Empathic Voice Interface (EVI) to deliver across three critical capabilities: Real Expressiveness: Their AI hosts need to not only read the script, but build a genuine relationship with their audience. EVI brings emotional intelligence and personality to every episode, transforming basic narration into an engaging conversation. Enterprise Scalability: Producing 3,000+ episodes weekly demands consistent, high-quality voices that don't degrade under volume. EVI maintains studio-grade audio fidelity whether Inception Point is creating 50 episodes or 50,000. Multi-Format Ready: As Inception Point expands beyond podcasting into short-form video, social media content, and interactive experiences, EVI provides voices that work across every medium. (337 words, Oct 10, 2025) - [Octave 2: next-generation multilingual voice AI | Hume Blog | Hume AI](/content/blog/octave-2-launch/index.html): Today we’re launching Octave 2, the second generation of our frontier voice AI model for text-to-speech. We just made a preview of Octave 2 available on our platform and through our API. (793 words, Oct 1, 2025) - [Hume AI powers conversational learning with Coconote | Hume Blog | Hume AI](/content/blog/case-study-hume-coconote/index.html): While traditional note-taking apps require students to manually scroll and search through content, Coconote is creating interactive study experiences through conversational AI. Coconote’s voice chat feature, powered by Hume's EVI, helps users transform static notes into dynamic conversations. Students can: Ask natural questions about their lecture content Receive contextual explanations referencing specific notes, and Engage in quiz-style conversations for active learning—all through natural voice interaction. (307 words, Sep 24, 2025) - [Preventing Harmful Deepfakes with Hume's EVI | Hume Blog | Hume AI](/content/blog/case-study-hume-reality-defender/index.html): Hume AI's partnership provided Reality Defender with strategic advantages through early access to our Empathic Voice Interface (EVI): Proactive Threat Intelligence: EVI helps Reality Defender understand and prepare for emerging voice synthesis capabilities before they become widely available, enabling next-generation audio detection that can identify sophisticated voice deepfakes. Comprehensive Dataset Generation: Reality Defender generates extensive datasets with EVI that refine and improve their audio detection models, with continuous algorithm refinement through their exposure to cutting-edge voice AI. (394 words, Aug 26, 2025) - [Democratizing Audiobook Creation with Hume's TTS | Hume Blog | Hume AI](/content/blog/case-study-spoken/index.html): Spoken is an innovative, AI-driven platform designed for independent authors who are seeking affordable alternatives to traditional audiobook production. Founded by Phil Marshall to solve the challenges he faced creating his own audiobook, Spoken empowers writers to transform their manuscripts into expressive, multi-cast audiobooks with full creative control. Powered by Hume's emotionally intelligent Text-to-Speech API, authors can now create expressive, multi-cast audiobooks that rival professional voice acting. (387 words, Aug 25, 2025) - [Hume AI brings expressive speech to Cerebras-powered Qwen reasoning models | Hume Blog | Hume AI](/content/blog/case-study-hume-cerebras/index.html): Hume AI has partnered with Cerebras to bring expressive, emotionally intelligent voice to Cerebras-hosted language models. The collaboration showcases Qwen-3-235b-a22b, a powerful reasoning model that demonstrates the synergy between Cerebras' ultra-fast inference infrastructure and Hume's Empathic Voice Interface (EVI), enabling voice agents that can think through complex problems while speaking with lifelike emotional intonation and word emphasis. (483 words, Jul 29, 2025) - [Hume AI brings expressive speech to Groq-powered Kimi K2 | Hume Blog | Hume AI](/content/blog/case-study-hume-groq/index.html): Hume AI has partnered with Groq to bring emotionally intelligent voice to Groq-hosted language models. The collaboration debuts with Kimi K2, supporting a multimodal assistant that showcases how Groq's LPU™ Inference Engine and Hume's Empathic Voice Interface (EVI) work together to create conversations that feel remarkably human. (564 words, Jul 29, 2025) - [Hume AI brings expressive speech to SambaNova-powered language models | Hume Blog | Hume AI](/content/blog/case-study-hume-sambanova/index.html): Hume AI has partnered with SambaNova to bring expressive, emotionally intelligent voice to SambaNova-hosted language models. The collaboration addresses a key need identified by SambaNova's extensive customer base, who have been seeking SambaNova-hosted TTS solutions to complement their existing LLM infrastructure, showcasing models like Llama-4-Maverick-17B-128E-Instruct and DeepSeek-R1-Distill-Llama-70B with seamless voice experiences within their trusted AI platform. (499 words, Jul 29, 2025) - [Vapi partners with Hume AI to offer developers the most affordable emotionally intelligent voice for phone calling | Hume Blog | Hume AI](/content/blog/case-study-hume-vapi/index.html): Vapi's platform serves developers building conversational voice agents in phone calling applications. Through a close partnership with Hume AI, they've integrated Hume's Octave TTS to offer developers real-time expressive text-to-speech at 150ms latency and 2¢/minute pricing, delivering significant performance and cost improvements over previous solutions. (376 words, Jul 29, 2025) - [Announcing EVI 3 API: The most customizable speech-to-speech model | Hume Blog | Hume AI](/content/blog/announcing-evi-3-api/index.html) (988 words, Jul 17, 2025) - [Expression Measurement for Emotional Intelligence | Hume Blog | Hume AI](/content/blog/case-study-hume-iris-insights/index.html): IRIS Insights (formerly EarningsEdge AI) is an AI-powered conversational intelligence platform helping organizations, from contact centers to correctional departments, turn human interactions into actionable emotional insight. Their mission is to decode not just what was said in a conversation, but how it was felt. By exclusively integrating Hume’s Expression Measurement API, IRIS adds a precise emotional layer to its language analysis stack. This allows teams to detect subtle changes in tone like frustration, confusion, or disengagement across every customer or inmate interaction. (391 words, Jun 18, 2025) - [Roleplays: How we wrote evals for the Hume MCP Server | Hume Blog | Hume AI](/content/blog/roleplays-evals-hume-mcp-server/index.html): We recently released the Hume AI MCP Server. This means if you that you can use Hume right from inside your AI chat using client applications — like Cursor or Claude Desktop — that implements the Model Context Protocol (MCP). You can use the AI to interactively help you with tasks like "can you help me narrate my audiobook?" or "can you help me design a voice for my video game character?". While it was previously possible to copy and paste between your favorite AI assistant and Hume, using Hume through MCP streamlines the process and can save you a lot of pointing and clicking. (2,135 words, Jun 10, 2025) - [Introducing EVI 3: the world’s most realistic and instructible speech-to-speech foundation model | Hume Blog | Hume AI](/content/blog/introducing-evi-3/index.html): At Hume, we promised ourselves that before the end of 2025, we’d achieve a voice AI experience that can be fully personalized. We believe this is an essential step toward voice being the primary way people want to interact with AI. (1,112 words, May 29, 2025) - [Voice AI for Interview Prep | Hume Blog | Hume AI](/content/blog/case-study-hume-interview-optimiser/index.html): Interview Optimiser is an AI-powered platform that helps job seekers prepare for interviews through realistic, voice-to-voice mock interviews. Their mission is to equip candidates with confidence and clarity by simulating high-pressure interview scenarios. By implementing Hume's Empathic Voice Interface (EVI), they deliver dynamic, adaptive conversations and provide feedback on vocal delivery—such as tone, confidence, and emotional regulation. Their customer base includes job seekers at all career stages, as well as recruiters looking to streamline early-stage screenings. This partnership has enabled them to offer a lifelike interview experience with actionable insights, bridging the gap between practice and real-world performance. (345 words, May 14, 2025) - [Voice AI for Social Anxiety Support | Hume Blog | Hume AI](/content/blog/case-study-hume-bearwith/index.html): Bearwith is an AI social anxiety coach that helps people improve their social skills in a judgment-free environment. Specializing in bite-sized lessons followed by practical role-playing exercises, Bearwith empowers users of all communication skill levels to practice difficult or stressful conversations without fear. By integrating Hume's Empathic AI, Bearwith has unlocked the ability to detect and respond to users' emotional states during practice sessions, providing a more supportive and personalized coaching experience that goes beyond traditional text-based interactions. (384 words, May 7, 2025) - [Expression Measurement for Voice Agent Testing | Hume Blog | Hume AI](/content/blog/case-study-hume-roark-ai/index.html): Roark AI provides a comprehensive platform for monitoring, evaluating, and simulating customer calls specifically for Voice AI developers. Their innovative solution eliminates the need for hours of manual testing by instantly simulating real customer scenarios and identifying potential failures before they reach customers. Serving companies across healthcare, customer support, and service-oriented businesses, Roark AI's mission is to help businesses proactively enhance their voice AI solutions to ensure high customer satisfaction ratings. (375 words, Apr 27, 2025) - [Voice AI for Music Support | Hume Blog | Hume AI](/content/blog/case-study-hume-jammy-chat/index.html): Jammy, a voice-first wellness app that uses emotionally attuned music to support users in real time, is creating an emotional operating system designed to help users feel seen, heard, and supported without the friction of traditional wellness tools. The platform combines conversational AI with emotionally aligned music recommendations, allowing users to speak freely like they would to a trusted friend, and receive playlists, prompts, or support tailored to how they actually feel. Hume’s EVI plays a critical role in this experience, enabling Jammy to analyze vocal cues such as tone, intensity, and nuance. This emotional signal powers Jammy’s real-time content recommendations, making each interaction feel personally relevant and emotionally attuned. (446 words, Apr 23, 2025) - [Emotion AI for Market Prediction | Hume Blog | Hume AI](/content/blog/case-study-hume-mood-metrics-ai/index.html): MoodMetrics AI is revolutionizing financial markets by decoding the emotional undercurrents in central bank communications. Founded in 2024 at the University of Bath, the company empowers financial institutions with emotional intelligence insights by analyzing verbal and non-verbal behavior, including vocal tone, speech patterns, and micro-expressions. By integrating Hume’s Expression Measurement API, MoodMetrics detects subtle emotional cues that traditional models miss—helping investment banks, retail banks, and hedge funds anticipate market movements with greater precision. The company has been nationally recognized through awards such as the Santander UK University Startup Grant and a Sparkies Award nomination. (310 words, Apr 15, 2025) - [The 8 Best AI Voice Generators in 2025 | Hume Blog | Hume AI](/content/blog/the-8-best-ai-voice-generators-in-2025/index.html): Artificial intelligence (AI) has ushered in groundbreaking innovations, and AI voice generators stand out as one of the most transformative. These tools use advanced text-to-speech and voice AI models to produce lifelike, human-sounding voices, unlocking endless possibilities across industries. From immersive video narrations and interactive e-learning modules to AI assistants that mimic real people, these generators are reshaping accessibility and content creation. If you're eager to explore this cutting-edge technology, here are the top 8 AI voice generators to watch in 2025. (665 words, Apr 11, 2025) - [Voice AI for Recruiting | Hume Blog | Hume AI](/content/blog/case-study-hume-nancy-ai/index.html): Nancy AI is a full AI recruiter that handles the hiring process end-to-end, utilizing over 100 AI agents for tasks ranging from job creation and candidate targeting to interviewing, offer generation, and onboarding. Nancy AI’s assessment services include competency-based interviews, situational judgment tests, and psychometric assessments, making it a comprehensive solution for both job seekers and businesses. At the core of Nancy AI’s voice interactions is Hume’s Empathic Voice Interface (EVI), which serves as the voice of Nancy, enabling emotionally intelligent conversations with candidates and employees. Nancy AI’s mission is twofold: for job seekers, it provides an interactive AI experience that simulates formal interviews while offering real-time feedback for personal development; for businesses, it delivers a comprehensive AI solution for end-to-end human capital management, streamlining recruitment and employee assessment. (319 words, Apr 9, 2025) - [How does Hume Octave compare to other leading TTS models like Elevenlabs? | Hume Blog | Hume AI](/content/blog/octave-tts-study-performance/index.html) (256 words, Apr 8, 2025) - [Voice AI for Mock Interviews | Hume Blog | Hume AI](/content/blog/case-study-hume-parrot-prep-ai/index.html): ParrotPrep is an AI-powered mock interview platform designed to help job seekers prepare for interviews through realistic practice sessions by offering customizable AI interviews across various career tracks, including Software Engineering, Data Science, Sales, and Business Consulting. Their mission is to eliminate interview anxiety through thorough preparation, guided by the principle: "If you fail to prepare, you are preparing to fail." (339 words, Apr 7, 2025) - [Voice AI for Language Learning | Hume Blog | Hume AI](/content/blog/case-study-hume-stimuler/index.html): Stimuler is a voice-first platform designed to help ESL (English as a Second Language) speakers globally improve their conversational skills. With over 3 million users worldwide and 1M+ downloads on Google Play alone, Stimuler has been recognized as the Best AI App by Google Play. Their mission is to provide the best product for ESL learners to practice and master English through immersive, voice-driven interactions. As a voice-first product, Stimuler’s users expect to engage in natural, flowing conversations to improve their skills. To enhance this experience, Stimuler experimented with a new conversational feature, made possible by Hume’s Empathic Voice Interface (EVI). This feature allows users to engage in empathic, voice-driven conversations, creating a more natural and engaging learning environment. (275 words, Apr 1, 2025) - [Voice AI for Financial Analysis | Hume Blog | Hume AI](/content/blog/case-study-hume-markets-eq/index.html): Markets EQ is a leading provider of performance intelligence tools designed to enhance financial decision-making through advanced voice, language, and emotional analysis. Specializing in the analysis of earnings calls and other critical financial communications, Markets EQ empowers investors, C-suite executives, and financial professionals to gain deeper preparatory support into high impact moments and critical communications. By integrating Hume’s Empathic AI, Markets EQ has unlocked the ability to capture and interpret emotional nuances in vocal communications, providing a more comprehensive understanding of financial data that goes beyond traditional metrics. (426 words, Mar 27, 2025) - [Voice AI for Customer Support | Hume Blog | Hume AI](/content/blog/case-study-hume-vonova/index.html): Customers find voice support more natural and reassuring—hearing a real voice instills trust and provides comfort that automated text responses often lack. As brands grow, they encounter a dilemma: while the demand for voice interactions rises, many are forced to mask their phone numbers due to high operational overheads. Vonova recognized this challenge and set out to redefine customer support. Vonova’s approach goes beyond simple automation. Their AI voice agents are designed to integrate flawlessly with existing systems while offering the empathy and adaptability that traditional call centers cannot match. The incorporation of Hume’s EVI was crucial to this transformation, adding a layer of emotional intelligence to every customer interaction. (309 words, Mar 25, 2025) - [Voice AI for Mental Health | Hume Blog | Hume AI](/content/blog/case-study-hume-hpy/index.html): hpy is committed to revolutionizing mental health care by integrating empathic AI into therapeutic practice. Their mission is to enhance the quality and accessibility of mental health support by reducing administrative burdens, providing real-time emotional insights during sessions, and extending therapeutic support between appointments. By harnessing Hume’s advanced technologies, hpy enables therapists to deliver more personalized and effective care while preserving the essential human connection that underpins successful therapy. At its core, hpy envisions a future where everyone has a personal wellbeing agent in their pocket—one that’s truly aligned with their emotional health and happiness, fostering genuine connections in a world where technology often extracts rather than enriches. (431 words, Mar 20, 2025) - [Voice AI for Mental Health Coaching | Hume Blog | Hume AI](/content/blog/case-study-hume-ream-app-limited/index.html): Ream is an innovative online mental health service dedicated to delivering effective and affordable support for adults with ADHD. The company faced challenges with low user retention and engagement using text-based tools and sought a solution that would add emotional nuance and personalization to its interactions. Recognizing that voice interactions feel more natural and empathetic, Ream set out to transform its coaching approach. Hume’s EVI has been a game-changer for Ream, enabling the company to deliver a coaching experience that truly resonates with users. (336 words, Mar 18, 2025) - [Emotion AI for Journaling | Hume Blog | Hume AI](/content/blog/case-study-hume-untold-app/index.html): Untold is on a mission to help people reconnect with their inner selves through audio journaling. Their app makes it easy and enjoyable to capture life’s moments while offering thoughtful prompts and insights to help users learn, grow, and reflect on their emotional journeys. By integrating Hume’s Expression Measurement API, Untold provides users with a detailed emotional readout of their journal entries, offering a moment of validation and reflection. This emotional analysis helps users feel seen and understood, fostering a deeper connection with their own experiences. (261 words, Mar 13, 2025) - [Voice AI for Consumer Research | Hume Blog | Hume AI](/content/blog/case-study-hume-university-of-zurich/index.html): Traditional AI voice assistants, like Alexa or Google Assistant, excel at functional tasks (e.g., setting reminders or finding information) but often lack the emotional depth needed for meaningful user connections, especially in emotionally driven settings like shopping for experiential products (e.g., scented candles) or customer service interactions. To address this gap, researchers from the University of Zurich (UZH) and ETH Zurich conducted a groundbreaking study as part of the AI Empathy Research Initiative, exploring how empathic AI could impact the consumer decision-making process by enhancing user satisfaction and emotional well-being. (428 words, Mar 10, 2025) - [Octave TTS Prompting Guide | Hume Blog | Hume AI](/content/blog/octave-tts-prompting-guide/index.html): While other text-to-speech models simply “read” words, Octave Text-to-Speech (TTS) is built on a language model, enabling it to interpret the meaning of text. With Octave, you can customize voices for any character, guide emotional delivery, and bring stories to life with human-like expression. The Octave speech-language model (speech LM) is a state-of-the-art voice AI model trained on data that captures the nuances of human vocal expression. It can interpret plot twists, emotional cues, and character traits within a script or prompt, transforming them into lifelike speech. To help you create the best possible samples and fully leverage the capabilities of this speech LM, we’ve compiled the following tips and tricks. (1,141 words, Feb 26, 2025) - [Octave TTS: the first text-to-speech system that understands what it’s saying | Hume Blog | Hume AI](/content/blog/octave-the-first-text-to-speech-model-that-understands-what-its-saying.html): Today we’re launching Octave (Omni-capable text and voice engine), the first LLM for text-to-speech. Unlike conventional TTS that merely “reads” words, Octave is a speech-language model that understands what words mean in context, unlocking a new level of expressiveness and nuance—and new AI voice capabilities. (1,457 words, Feb 26, 2025) - [Voice AI for Automotive UX | Hume Blog | Hume AI](/content/blog/case-study-hume-automotive/index.html): In collaboration with Hume, a Fortune 100 automotive company conducted a study to explore how drivers interact with emotionally intelligent AI in vehicles. Using Hume’s Empathic Voice Interface (EVI), the study aimed to understand how an AI that responds to users’ emotions could create richer, more personalized driving experiences. Over four weeks, a diverse group of drivers engaged with EVI while driving. They asked for directions, checked the weather, and even discussed fun facts. By the end of the study, it became clear that drivers preferred voice assistants with more personality over those focused solely on utility. (415 words, Feb 19, 2025) - [Voice AI for Thought Leadership | Hume Blog | Hume AI](/content/blog/case-study-hume-pressmaster/index.html): In the day and age of social media marketing, content creation from company leaders has emerged as an important pathway to highlighting thought leadership and building personal brand and company credibility. However, traditional content creation methods are time-consuming, requiring extensive writing and editing, while simpler automation tools often fail to capture authentic voice and nuanced insights. Recognizing these challenges, Pressmaster.AI developed an AI-powered interview system that transforms casual conversations between individuals and AI into polished thought leadership content. To create an engaging and natural conversation that truly captures users' insights and personality, they needed a voice AI solution capable of conducting interviews with the emotional intelligence of a skilled journalist. They found their answer in Hume's Empathic Voice Interface (EVI). (393 words, Feb 7, 2025) - [Best ways to talk to AI with voice: NotebookLM, Advanced Voice Mode, and EVI 2 | Hume Blog | Hume AI](/content/blog/best-ways-to-talk-to-ai-with-voice/index.html): Artificial intelligence (AI) is revolutionizing how we interact with technology, and voice-based communication is at the forefront of this transformation. From personalized AI assistants to emotionally intelligent conversational agents, voice-enabled AI is becoming more sophisticated and human-like. In this article, we’ll explore three cutting-edge voice AI technologies: Google’s NotebookLM, OpenAI’s Advanced Voice Mode, and Hume AI’s EVI 2. Each of these platforms offers unique capabilities that redefine how we communicate with AI. (878 words, Jan 28, 2025) - [Speech-to-text and text-to-speech | Hume Blog | Hume AI](/content/blog/speech-to-text-and-text-to-speech-stt-tts/index.html): Speech-to-text (STT) and text-to-speech (TTS) are two revolutionary technologies that have fundamentally changed how we interact with computers and digital devices. Major tech companies like Google, IBM, and Amazon are in a constant race to develop the most accurate and advanced speech recognition systems. While both STT and TTS involve converting between spoken and written language, they serve different purposes and are used in a variety of applications. This article will explore how each technology works, examine their wide-ranging uses, analyze their strengths and limitations, and highlight the latest advancements and future trends in the field. (1,667 words, Jan 28, 2025) - [Speech-language models: A deeper dive into voice AI | Hume Blog | Hume AI](/content/blog/speech-language-models-a-deeper-dive-into-voice-ai/index.html): Speech-language models are set to revolutionize voice AI, offering a level of sophistication and nuance that surpasses traditional technologies. These models don’t just process speech—they understand it, capturing the subtleties of human communication, from tone and emotion to context and intent. This article explores how speech-language models like EVI 2, Moshi, and GPT-4o are redefining voice AI, and what their advancements mean for the future of human-computer interaction. (1,016 words, Jan 27, 2025) - [Controlling your computer with voice: Windows, MacOS, and AI Solutions | Hume Blog | Hume AI](/content/blog/controlling-your-computer-with-voice/index.html): Imagine seamlessly controlling your computer with just your voice, like a scene from a sci-fi movie. You could launch applications, compose emails, and navigate the web—all without touching a keyboard or mouse. This is the transformative power of voice control, a technology revolutionizing how we engage with our devices. Beyond enhancing accessibility for individuals with disabilities, voice control offers a more efficient and intuitive way for everyone to interact with their computers. (2,044 words, Jan 22, 2025) - [EVI 2 + Claude Computer Use | Hume Blog | Hume AI](/content/blog/evi2-claude-computer-use/index.html): You can now control a computer with just your voice. In just a few hours, we combined the EVI 2 API with Claude's new computer use functionality. Here’s how it works. (185 words, Jan 22, 2025) - [How to generate AI voices with accents | Hume Blog | Hume AI](/content/blog/how-to-generate-ai-voices-with-accents/index.html): The use of AI voices is rapidly increasing in various applications, including voice assistants, audiobooks, and video games. One of the key features of AI voices is the ability to generate different accents, which can enhance the user experience and create more realistic and engaging content. This article explores how to generate AI voices with accents, the different types of accents available, the challenges and limitations, and the ethical considerations involved. (1,655 words, Jan 17, 2025) - [How to tell human voices from AI | Hume Blog | Hume AI](/content/blog/how-to-tell-human-voices-from-ai/index.html): The rise of artificial intelligence (AI) has brought about incredible advancements in various fields, and one area where its impact is particularly noticeable is voice generation. AI-powered voice generators can now create synthetic voices that are remarkably close to human speech, making it increasingly difficult to distinguish between the two. This has significant implications for various applications, from interactive voice response systems and virtual assistants to audiobooks and podcasts. However, despite the impressive progress, there are still ways to tell human voices from AI-generated ones. (1,989 words, Jan 17, 2025) - [How to create voiceovers for YouTube videos | Hume Blog | Hume AI](/content/blog/how-to-create-voiceovers-for-youtube-videos/index.html): Creating high-quality voiceovers is essential for engaging viewers and enhancing the overall quality of your YouTube videos. Whether you're creating tutorials, explainer videos, or vlogs, a well-produced voiceover can make all the difference. This comprehensive guide will walk you through the process of creating voiceovers for your YouTube videos, covering everything from equipment and software to recording and editing techniques. (1,540 words, Jan 15, 2025) - [How to create a faceless YouTube channel | Hume Blog | Hume AI](/content/blog/how-to-create-a-faceless-you-tube-channel/index.html): The rise of faceless YouTube channels is a growing trend, allowing creators to share content and build an audience without revealing their identity. This approach offers privacy and allows creators to prioritize their content over personal appearance. However, with the increasing accessibility of AI tools, the faceless YouTube niche is becoming more competitive. To succeed, creators need to produce unique, high-quality content that stands out. This article explores how to create a faceless YouTube channel using an AI voice model, covering everything from video format selection to channel optimization and monetization strategies. (1,679 words, Jan 15, 2025) - [Creating custom character voices with AI | Hume Blog | Hume AI](/content/blog/creating-custom-character-voices-with-ai/index.html): The use of AI in voice generation is rapidly changing how we interact with technology and consume content. From video games and animated films to audiobooks and virtual assistants, AI-generated voices are becoming increasingly prevalent. One of the most exciting applications of this technology is in the creation of custom character voices. This allows developers and creators to imbue their characters with unique and engaging personalities, enhancing the overall user experience. This article delves into the best ways to create custom character voices using AI, exploring the various technologies, platforms, and customization options available. (1,451 words, Jan 14, 2025) - [Creating video game character voices with AI | Hume Blog | Hume AI](/content/blog/creating-video-game-character-voices-with-ai/index.html): The video game industry is constantly evolving, with developers always seeking new ways to enhance immersion and create dynamic gaming experiences. One of the key areas of focus is character voices. Traditionally, giving characters a voice involved hiring voice actors to record lines of dialogue, which could be expensive and time-consuming. However, advancements in artificial intelligence (AI) have introduced innovative solutions for generating character voices, offering game developers more efficient and flexible options. This article explores how AI can be used to create video game character voices, examining the processes, platforms, and potential benefits and drawbacks. (1,547 words, Jan 14, 2025) - [Designing custom voices with AI | Hume Blog | Hume AI](/content/blog/designing-custom-voices-with-ai/index.html): The use of artificial intelligence (AI) in voice design is revolutionizing how we interact with technology and experience audio content. AI-powered systems can create synthetic voices with specific characteristics, clone the voices of real people, and even generate entirely new voices with unique qualities. This has led to a surge in creative applications, such as producing realistic voiceovers for films and video games, developing personalized voices for virtual assistants, and restoring the voices of individuals with speech impairments. This article explores the current landscape of AI-powered voice design systems, examining the technologies, capabilities, ethical considerations, and potential of this rapidly evolving field. We will delve into the work of various companies and research labs at the forefront of this innovation, including ElevenLabs, Respeecher, Microsoft, Google, AWS, MIT CSAIL, and Hume AI. (1,683 words, Jan 14, 2025) - [How to clone your voice with AI | Hume Blog | Hume AI](/content/blog/how-to-clone-your-voice-with-ai/index.html): Voice cloning technology has advanced significantly in recent years, making it possible to create a synthetic replica of your voice with remarkable accuracy. This technology has various applications, from entertainment and personalized content creation to accessibility and assistive technologies. This article explores the process of cloning your voice with AI, examining the steps involved, the tools available, and the ethical and legal considerations. (1,598 words, Jan 14, 2025) - [Speech-to-text and text-to-speech | Hume Blog | Hume AI](/content/blog/speech-to-text-and-text-to-speech/index.html): Speech-to-text (STT) and text-to-speech (TTS) are two groundbreaking technologies that have transformed how we engage with computers and other devices. Leading tech companies like Google, IBM, and Amazon are constantly competing to develop the most accurate and sophisticated speech recognition systems. While both STT and TTS involve converting between spoken and written language, they have distinct functions and applications. This article explores the inner workings of each technology, examines their diverse use cases, analyzes their strengths and weaknesses, and discusses the current advancements and future trends in the field. (1,269 words, Jan 14, 2025) - [Creating AI voiceovers with emotion | Hume Blog | Hume AI](/content/blog/creating-ai-voiceovers-with-emotion/index.html): The use of artificial intelligence (AI) in voice generation has progressed significantly, moving beyond the monotonous, robotic voices of the past to more natural and expressive speech. This evolution has opened up exciting possibilities for various applications, including voiceovers. This article delves into current techniques for creating AI voiceovers with nuanced emotions like anger, sadness, or calmness, explores the upcoming capabilities of next-generation AI voice models like Hume AI's OCTAVE, and discusses the potential benefits and challenges of this technology. (1,129 words, Jan 14, 2025) - [Controlling the speed of AI voices | Hume Blog | Hume AI](/content/blog/controlling-the-speed-of-ai-voices/index.html): The rise of artificial intelligence has revolutionized many aspects of our lives, and one area where its impact is increasingly felt is in the realm of voice technology. AI voices are now commonly used in virtual assistants, customer service bots, audiobooks, video games, and various accessibility tools. As this technology continues to evolve, a key question emerges: to what extent can we control the speed (speech rate) of these synthetic voices? (1,014 words, Jan 13, 2025) - [Introducing OCTAVE (Omni-Capable Text and Voice Engine) | Hume Blog | Hume AI](/content/blog/introducing-octave/index.html): A frontier speech-language model with new emergent capabilities, like on-the-fly AI voice and personality creation. (843 words, Dec 23, 2024) - [Introducing Voice Control | Hume Blog | Hume AI](/content/blog/introducing-voice-control/index.html): We’re introducing Voice Control, a novel interpretability-based method that brings precise control to AI voice customization without the risks of voice cloning. Our tool gives developers control over 10 voice dimensions, labeled “masculine/feminine,” “assertiveness,” “buoyancy,” “confidence,” “enthusiasm,” “nasality,” “relaxedness,” “smoothness,” “tepidity,” and “tightness.” Unlike prompt-based approaches, Voice Control enables continuous adjustments along these dimensions, allowing for precise control and making voice modifications reproducible across sessions. (732 words, Dec 2, 2024) - [Hume AI + Anthropic create emotionally intelligent voice interactions | Hume Blog | Hume AI](/content/blog/hume-anthropic-claude-voice-interactions/index.html): Hume AI trained its speech-language foundation model to verbalize Claude responses, powering natural, empathic voice conversations that help developers build trust with users in healthcare, customer service, and consumer applications. (1,053 words, Nov 22, 2024) - [Voice AI for Eldercare | Hume Blog | Hume AI](/content/blog/case-study-hume-everfriends/index.html): To truly connect with users and provide a natural, empathic experience, EverFriends.ai needed an AI solution capable of understanding and responding to emotional cues. They found their answer in Hume's Empathic Voice Interface (EVI). EVI merges generative language and voice into a single model trained specifically for emotional intelligence, enabling it to emphasize the right words, laugh or sigh at appropriate times, and much more, guided by language prompting to suit any particular use case. (525 words, Nov 4, 2024) - [How can emotionally intelligent voice AI support our mental health? | Hume Blog | Hume AI](/content/blog/voice-ai-mental-health/index.html): Recent advances in voice-to-voice AI, like EVI 2, offer emotionally intelligent interactions, picking up on vocal cues related to mental and physical health, which could enhance both clinical care and daily well-being. (1,632 words, Oct 22, 2024) - [Are emotional expressions universal? | Hume Blog | Hume AI](/content/blog/are-emotion-expressions-universal/index.html): Do people around the world express themselves in the same way? Does a smile mean the same thing worldwide? And how about a chuckle, a sigh, or a grimace? These questions about the cross-cultural universality of expressions are among the more important and long-standing in behavioral sciences like psychology and anthropology—and central to the study of emotion. (1,340 words, Oct 4, 2024) - [Can AI “detect” emotions? | Hume Blog | Hume AI](/content/blog/can-ai-detect-emotions/index.html): For AI to enhance our emotional well-being and engage with us meaningfully, it needs to understand the way we express ourselves and respond appropriately. This capability lies at the heart of a field of AI research that focuses on machine learning models capable of identifying and categorizing emotion-related behaviors. However, this area of research is frequently misunderstood, often sensationalized under the umbrella term "emotion AI"--AI that can “detect” emotions, an impossible form of mind-reading. (1,272 words, Sep 23, 2024) - [What is emotion science? | Hume Blog | Hume AI](/content/blog/what-is-emotion-science/index.html): How can artificial intelligence achieve the level of emotional intelligence required to understand what makes us happy? As AI becomes increasingly integrated into our daily lives, the need for AI to understand emotional behaviors and what they signal about our intentions and preferences has never been more critical. (1,043 words, Sep 23, 2024) - [Comparing the world’s first voice-to-voice AI models: EVI 2 and GPT-4o | Hume Blog | Hume AI](/content/blog/evi2-vs-gpt4ovoice/index.html): The world’s first working voice-to-voice models are Hume AI's Empathic Voice Interface 2 (EVI 2) and OpenAI's GPT-4o Advanced Voice Mode. EVI 2 is publicly available, as an app and an API that developers can build on. GPT-4o voice is available as the OpenAI Realtime API. Here we explore the similarities, differences, and potential applications of these systems. (1,800 words, Sep 11, 2024) - [Introducing EVI 2, our new foundational AI voice model | Hume Blog | Hume AI](/content/blog/introducing-evi2/index.html): EVI 2 is our new foundational voice-to-voice model. It is one of the first AI models with which you can have remarkably human-like voice conversations. It can converse rapidly and fluently with users with subsecond response times, understand a user’s tone of voice, generate any tone of voice, and even respond to some more niche requests like changing its speaking rate or rapping. It can emulate a wide range of personalities, accents, and speaking styles and possesses emergent multilingual capabilities. (481 words, Sep 11, 2024) - [Emotion AI for Personal Growth Companions | Hume Blog | Hume AI](/content/blog/case-study-hume-new-computer/index.html): Dot, a personal AI developed by New Computer, is designed to grow alongside its users, serving as a thought partner, friend, and confidant for young adults navigating uncertainty in their lives. Built on an infinite long-term memory system, Dot's responses become increasingly personalized with each interaction, whether through speaking or typing. (417 words, Aug 22, 2024) - [Emotion AI for Language Learning Toys | Hume Blog | Hume AI](/content/blog/case-study-hume-robotics/index.html): A leading Japanese robotics company sought to develop a conversational stuffed animal to enhance English language learning for Japanese children. (339 words, Aug 22, 2024) - [Voice AI for Preventative Healthcare | Hume Blog | Hume AI](/content/blog/case-study-hume-thumos/index.html): Thumos Care, founded by an experienced startup operator and board-certified physician duo, developed a solution that leverages conversational AI to bridge the gap between doctor visits and provide continuous patient care. (480 words, Aug 22, 2024) - [Emotion AI for Audience Growth | Hume Blog | Hume AI](/content/blog/case-study-hume-tone-ai/index.html): How Tone AI uses Hume’s Expression Measurement API to boost audience growth for NFL teams and media organizations (329 words, Aug 22, 2024) - [Publication in Frontiers in Psychology: Insights from a Large-Scale Study on the Meanings of Facial Expressions Across Cultures | Hume Blog | Hume AI](/content/blog/large-study-facial-expressions/index.html): Understanding how emotions are experienced and expressed across different cultures has long been a central focus of debate and study in psychology, cognitive science, and anthropology. What emotions do people in different cultures experience in response to the same evocative scenes and scenarios? What facial movements do they produce? How are feelings and expressions related? (1,158 words, Jun 28, 2024) - [EVI Web Search Demo: The First Interactive Voice AI Podcast | Hume Blog | Hume AI](/content/blog/chatter-interactive-podcast/index.html): Hume’s Empathic Voice Interface (EVI) is now the first voice AI API capable of native web search. (188 words, May 15, 2024) - [Introducing Hume’s Empathic Voice Interface (EVI) API | Hume Blog | Hume AI](/content/blog/introducing-hume-evi-api/index.html): Last month, we released the demo of our Empathic Voice Interface (EVI). The first emotionally intelligent voice AI API is finally here! EVI does a lot more than stitch together transcription, LLMs, and text-to-speech. With a new empathic LLM (eLLM) that processes your tone of voice, EVI unlocks new capabilities like knowing when to speak, generating more empathic language, and intelligently modulating its own tune, rhythm, and timbre. EVI is the first voice AI that really sounds like it understands you. (685 words, Apr 18, 2024) - [Hume Raises $50M Series B and Releases New Empathic Voice Interface | Hume Blog | Hume AI](/content/blog/series-b-evi-announcement/index.html): To build emotionally intelligent voice AI, Hume AI raised a $50m Series B round led by EQT Ventures and joined by Union Square Ventures, Nat Friedman & Daniel Gross, Metaplanet, Northwell Holdings, Comcast Ventures, and LG Technology Ventures. (740 words, Mar 25, 2024) - [What is semantic space theory? | Hume Blog | Hume AI](/content/blog/what-is-semantic-space-theory/index.html): Our models and products are built on a cutting-edge approach to understanding emotion: semantic space theory (SST), which uses computational methods and data-driven approaches to map the full spectrum of our feelings (1,238 words, Feb 21, 2024) - [Publication in iScience: Understanding what facial expressions mean in different cultures | Hume Blog | Hume AI](/content/blog/iscience-facial-expression-different-culture/index.html): How many different facial expressions do people form? How do they differ in meaning across cultures? Can AI capture these nuances? Our new paper provides new in-depth answers to these questions with the help of machine learning. (1,520 words, Feb 20, 2024) - [Introducing a new evaluation for creative ability in Large Language Models | Hume Blog | Hume AI](/content/blog/new-evaluation-creative-ability-in-large-language-models.html): Introducing HumE-1 (Human Evaluation 1), our new evaluation for large language models (LLMs) that uses human ratings to evaluate LLMs for their ability to perform creative tasks in the ways that matter to us, evoking the feelings we want to feel. (973 words, Feb 9, 2024) - [Tutorial: Hands-on with Hume's Custom Model API | Hume Blog | Hume AI](/content/blog/hume-custom-model-api/index.html): Our Custom Model API can be used to predict your own labels of any audio, video, image, or text file. In this short tutorial, we'll walk you through how to build a Custom Model that integrates language, voice, and/or facial movement and is automatically deployed through our API. (1,348 words, Dec 18, 2023) - [Announcing our Custom Model API | Hume Blog | Hume AI](/content/blog/announcing-our-custom-model-api/index.html): Hume's Custom Model API integrates dynamic patterns of facial expression, vocal expression, and/or language into a custom multimodal model, leveraging insights gained by pretraining on millions of videos and audio files to predict your labels as accurately as possible, even after seeing just a few dozen new examples. (423 words, Dec 14, 2023) - [Hume Powers Projects at UC Berkeley LLM Hackathon | Hume Blog | Hume AI](/content/blog/hume-powers-projects-at-uc-berkeley-llm-hackathon/index.html): UC Berkeley hosted the world's largest AI hackathon on June 17-18, with over 1200 students developing diverse applications using large language models and open-source APIs. Hume's APIs were used alongside LLMs in 57 of the 240 projects and three of the 12 finalists, ranging from healthcare devices to education tools, indicating its wide applicability and versatility. (477 words, Aug 11, 2023) - [Hume AI Raises $12.7M in Series A Funding | Hume Blog | Hume AI](/content/blog/hume-ai-raises-usd12-7m-in-series-a-funding/index.html): We're pleased to announce that we've raised a $12.7M Series A led by Union Square Ventures! We're excited to be partnering with leading investors who understand the importance of technology that can align itself with human well-being using the same cues humans do: our expressions. (645 words, Feb 16, 2023) - [Hume AI Publication in Nature Human Behavior: Deep Learning & Vocal Bursts in Different Cultures | Hume Blog | Hume AI](/content/blog/hume-ai-publication-in-nature-human-behavior-deep-learning-and-vocal-bursts.html): Our first in a series of publications is in Nature Human Behavior! "Deep learning reveals what vocal bursts express in different cultures" addresses a key question at the intersection of AI and psych: What does the voice convey without words? And do chuckles, gasps, sighs, etc. have the same meaning across cultures? (1,536 words, Jan 11, 2023) - [Tutorial: Hands-on with Hume AI’s API | Hume Blog | Hume AI](/content/blog/tutorials-hume-api/index.html): Working with the Hume AI Platform is easy! Built atop rigorous scientific studies of human expressive behavior, it's the only toolkit you need to measure nonverbal cues in audio, video, and images. Get started with our API, and integrate our cutting-edge models into your applications. (958 words, Sep 9, 2022) - [Episode 25 Myths About Emotion Science | The Feelings Lab | Hume Blog | Hume AI](/content/blog/the-feelings-lab-ep-25-myths-about-emotion-science/index.html): Do expressions “reveal our emotions”? What did Darwin think? Is “emotion AI” really what it sounds like? In this episode, Dacher Keltner helps us unpack popular myths and misconceptions about emotion science, including how recent headlines have departed from peer-reviewed science. (446 words, Aug 30, 2022) - [Can AI Teach Itself to Improve Our Well-Being? | Hume Blog | Hume AI](/content/blog/can-ai-teach-itself-to-improve-our-well-being/index.html): The future of technology hinges on the measurement of well-being, the importance of which can’t be overstated. As emotion scientists work with AI practitioners to translate this knowledge into ML models, we ask: Can AI teach itself to make the world a better, happier place? (1,464 words, Aug 9, 2022) - [Welcome to the Hume AI Blog | Hume Blog | Hume AI](/content/blog/welcome-to-the-hume-ai-blog/index.html): Explore empathic AI, emotion science, and voice technology on the Hume AI blog. Discover cutting-edge research, product updates, and developer guides for emotionally intelligent AI. (976 words, Aug 3, 2022) - [Episode 24 Trust & Safety in Online Conversations | The Feelings Lab | Hume Blog | Hume AI](/content/blog/the-feelings-lab-ep-24-trust-and-safety-in-online-conversations.html): Amid rising concerns over bots, BrandBastion CEO Jenny Wolfram joins us to discuss how AI is also empowering authentic voices, from classifying harmful content to delivering genuine feedback to organizations. (383 words, Jul 19, 2022) - [Episode 24 Compassion & Customer Service | The Feelings Lab | Hume Blog | Hume AI](/content/blog/the-feelings-lab-ep-24-compassion-and-customer-service-w-josh-feast.html): In a world of products, customer service is essential to our well-being. In this episode, Cogito CEO Josh Feast joins us to discuss how companies are studying nonverbal behavior to build more empathy into customer service. (210 words, Jun 14, 2022) - [Episode 23 Well-being | The Feelings Lab | Hume Blog | Hume AI](/content/blog/the-feelings-lab-ep-23-the-feelings-lab-well-being-w-dacher-keltner.html): Can AI teach itself to improve our well-being? Are tech companies really maximizing engagement at all costs? Is Gen Z okay? In this episode, Dacher Keltner joins us to discuss technology, well-being measurement, and implications for AI ethics. (234 words, Jun 7, 2022) - [Episode 22 Listener Questions + Emotion Science News | The Feelings Lab | Hume Blog | Hume AI](/content/blog/the-feelings-lab-ep-22-listener-questions-emotion-science-news.html): In this episode, we address our best pod-listener questions: Do video calls suppress creativity? Is expressing your emotions good for your mental health? Is there a Western bias in emotion science? Do lobsters have emotions? (411 words, May 31, 2022) - [Episode 21 Pain and Personalized Medicine | The Feelings Lab | Hume Blog | Hume AI](/content/blog/the-feelings-lab-ep-21-pain-and-personalized-medicine.html): Pain is subjective, but its diagnosis can have life-altering implications. Dr. Daniel Barron, Harvard psychiatrist and Director of Pain Intervention & Digital Research, joins us to discuss how objective measures of emotional expression can improve outcomes for patients. (330 words, May 10, 2022) - [Episode 20 Empathy and User Research | The Feelings Lab | Hume Blog | Hume AI](/content/blog/the-feelings-lab-ep-20-empathy-and-user-research/index.html): How can we foster empathy with users at the speed and scale needed to drive innovation in AI? Michael Winnick, CEO of dscout, joins us to discuss how empathic technologies can help researchers design products and user experiences. (302 words, Apr 27, 2022) - [Episode 19 ICML Expressive Vocalization Competition Panel | The Feelings Lab | Hume Blog | Hume AI](/content/blog/the-feelings-lab-ep-19-special-conversation-icml-exvo-2022-workshop-and.html): Vocalizations like laughs and sighs provide avenues to optimize AI for our well-being. Dr. Kory Mathewson, Dr. Gauthier Gidel, Dr. Panagiotis Tzirakis, and Dr. Alice Baird join us to discuss expressive vocalizations and machine learning. (305 words, Apr 17, 2022) - [Episode 18 Aesthetic Appreciation and Fine Art | The Feelings Lab | Hume Blog | Hume AI](/content/blog/the-feelings-lab-ep-18-aesthetic-appreciation-and-fine-art.html): How close is AI to understanding art? Kathy Tafel, Senior Director of Engineering at Artsy—the world's largest online art marketplace—joins us to discuss how AI tools for artists, curators, brokers, and more are democratizing the world of fine art. (309 words, Mar 29, 2022) - [Episode 17 Empathy and Augmented Reality | The Feelings Lab | Hume Blog | Hume AI](/content/blog/the-feelings-lab-ep-17-empathy-and-augmented-reality.html): Mursion CTO Arjun Nagendran joins us to discuss how augmented reality can help humans arrive at better decisions and more personal connections, bringing more empathy to interactions across neurodiverse individuals, cultures, and digital platforms. (376 words, Mar 22, 2022) - [Episode 16 Empathy and Digital Health | The Feelings Lab | Hume Blog | Hume AI](/content/blog/the-feelings-lab-ep-16-empathy-and-digital-health/index.html): Stanford professor and Cognoa co-founder Dennis Wall joins us to discuss the promise of AI in medicine, from democratizing the diagnosis of autism to enabling fine-grained dimensional assessments of pain, depression, and more. (395 words, Mar 8, 2022) - [Episode 15 Awe and Digital Art | The Feelings Lab | Hume Blog | Hume AI](/content/blog/the-feelings-lab-ep-15-awe-and-digital-art/index.html): NVIDIA VP Richard Kerris joins us to discuss how AI is transforming art and entertainment, helping artists create ground-breaking extensions of reality and human imagination that evoke greater and more frequent experiences of awe and the sublime. (401 words, Mar 1, 2022) - [Episode 14 Love and the Future of Dating | The Feelings Lab | Hume Blog | Hume AI](/content/blog/the-feelings-lab-ep-14-love-and-the-future-of-dating.html): AI is sourcing our dates. But is it really helping us find love? In this episode, we're joined by Kellie Ammerman, president of the matchmaking app Tawkify, as we discuss how technology might help us form deeper connections and create more love in the world. (273 words, Feb 22, 2022) - [Episode 13 Well-being in a Remote World | The Feelings Lab | Hume Blog | Hume AI](/content/blog/the-feelings-lab-ep-13-well-being-in-a-remote-world.html): In a world where most of our social interactions are remote and automated systems increasingly orchestrate our digital lives, Harris Poll CEO John Gerzema joins us to discuss how we can measure well-being and ensure people are cared for. (243 words, Feb 14, 2022) - [Episode 12 Fulfillment in the Metaverse | The Feelings Lab | Hume Blog | Hume AI](/content/blog/the-feelings-lab-ep-12-fulfillment-in-the-metaverse.html): We're joined by the co-founder of Soul Machines, Greg Cross, to discuss how we can build a more fulfilling digital world, addressing the promises and pitfalls of AI, digital people, and the metaverse. (405 words, Feb 8, 2022) - [Episode 11 Compassion and Robots | The Feelings Lab | Hume Blog | Hume AI](/content/blog/the-feelings-lab-ep-11-compassion-and-robots/index.html): In the Season 2 premiere of The Feelings Lab, we're joined by the CEO of Embodied, Dr. Paolo Pirjanian, as we discuss the future of robots, the prospects and pitfalls of empathic AI, and what really separates WALL-E from HAL-9000. (316 words, Feb 1, 2022) - [A Potentially More Humane Way to Let the Machines In | Hume Blog | Hume AI](/content/blog/the-washington-post-former-google-scientist/index.html): The Post writes that we bring a scientific approach to aligning AI with well-being, with "a high degree of psychology research to accompany those ethical goals" and "an ethics committee with many heavy-hitters in the field of emotional and ethical AI." (13 words, Jan 17, 2022) - [Episode 10 Holiday Emotions | The Feelings Lab | Hume Blog | Hume AI](/content/blog/the-feelings-lab-ep-10-holiday-emotions/index.html): In this episode, we’re joined by filmmaker, futurist, and public speaker Jason Silva as we discuss the bittersweet feelings of nostalgia, contemplation, and satisfaction associated with the holiday season. (391 words, Dec 14, 2021) - [Episode 9 Love | The Feelings Lab | Hume Blog | Hume AI](/content/blog/the-feelings-lab-ep-9-love/index.html): In this episode, we discuss the science of love with special guest, actor Melina Kanakeredes (“CSI: NY” and “The Resident”), who teaches us eight Greek words denoting different types of love (that the science of emotion might want to adopt). (311 words, Nov 23, 2021) - [Episode 8 Doubt | The Feelings Lab | Hume Blog | Hume AI](/content/blog/the-feelings-lab-ep-8-doubt/index.html): The Feelings Lab hosts discuss Dr. Alan Cowen’s favorite emotion: doubt. This is more than just the dark side of curiosity. In fact, it may be the most pivotal emotion in human history. (273 words, Nov 16, 2021) - [Episode 7 Mirth | The Feelings Lab | Hume Blog | Hume AI](/content/blog/the-feelings-lab-ep-7-mirth/index.html): In this episode, we discuss the emotion of mirth with best-selling author and comedian John Hodgman. After learning how this ancient emotion, shared by all mammals, has evolved from horseplay into human wordplay, you’ll never look at laughter the same way. (273 words, Nov 9, 2021) - [Episode 6 Anxiety | The Feelings Lab | Hume Blog | Hume AI](/content/blog/the-feelings-lab-ep-6-anxiety/index.html): In this episode, we discuss the emotion of anxiety with special guest Dr. George Bonanno, a world-renowned expert in the field of bereavement and trauma, and learn how to cope with this all-too-pervasive feeling. (197 words, Nov 2, 2021) - [Episode 5 Horror | The Feelings Lab | Hume Blog | Hume AI](/content/blog/the-feelings-lab-ep-5-horror/index.html): Award-winning actor and comedian Fred Armisen joins us to discuss the paradoxes of horror, the epitome of all negative emotion—a mix of disgust, fear, and existential dread—that, when experienced safely, we strangely enjoy. (305 words, Oct 26, 2021) - [Episode 4 Pride | The Feelings Lab | Hume Blog | Hume AI](/content/blog/the-feelings-lab-ep-4-pride/index.html): With professional volleyball player Cassidy Lichtman, we discuss pride, an emotion that drives human achievement and structures the social hierarchy, but clearly also has a dark side. (306 words, Oct 19, 2021) - [Episode 3 Desire | The Feelings Lab | Hume Blog | Hume AI](/content/blog/the-feelings-lab-ep-3-desire/index.html): In our third episode, our hosts and drag queen Monet X Change discuss the vast spectrum of desire, the inherent politics and mysticism embedded in the emotion, its evolution, and its impact on culture.  (302 words, Oct 11, 2021) - [Episode 2 Embarrassment | The Feelings Lab | Hume Blog | Hume AI](/content/blog/the-feelings-lab-ep-2-embarrassment/index.html): In our second episode, we discuss the feeling of embarrassment with comedian Ali Kolbert and Dr. Jessica Tracy. It may be unpleasant, but it’s glue for the fabric of our social lives.  (294 words, Oct 4, 2021) - [Episode 1 Awe | The Feelings Lab | Hume Blog | Hume AI](/content/blog/the-feelings-lab-ep-1-awe/index.html): In Episode 1, we explore the emotion that challenges us to confront vastness and mystery. Special guest Tami Simon, founder of Sounds True, weighs in. (274 words, Sep 27, 2021) - [Episode 0 Intro | The Feelings Lab | Hume Blog | Hume AI](/content/blog/the-feelings-lab-intro/index.html): An introduction to The Feelings Lab. Meet our hosts and step into a new era of science and technology that accounts for the complexity of human emotion. (464 words, Sep 26, 2021) ## About Pages - [About Hume AI | Hume AI](/content/about/index.html): Hume AI is building the data and evaluation layer for emotionally intelligent voice AI: scientifically grounded datasets, speech models, evaluation frameworks, and human feedback pipelines. (540 words) - [Contact Hume AI - Get in Touch | Hume AI](/content/contact/index.html): Contact Hume AI for sales inquiries, developer support, academic partnerships, or press inquiries. We'd love to hear from you. (159 words) ## Resources - [Full Page Index](/index.html): Browse all cached pages with rich metadata - [About This Cache](/content/about.html): Methodology, technical details, and usage guidelines - [XML Sitemap](/sitemap.xml): Machine-readable sitemap for crawler discovery - [Robots.txt](/robots.txt): Crawler directives