hume-evi
Build production realtime voice agents with Hume's Empathic Voice Interface (EVI): speech-to-speech sessions, empathic/prosodic response design, EVI 3 versus EVI 4-mini selection, WebSocket and SDK integration, tools/function calling, interruption/barge-in, turn-taking, transcripts/audio artifacts, pricing/limits, privacy, safety, consent, and QA.
Profile →npx skills add calesthio/generative-media-skills --skill hume-evi --agent claude-code
Same command for any agent — swap --agent for codex, cursor, copilot.
Weekly change comes from our own snapshots, not the repository page — it measures attention, not adoption.
What it does
Design, implement, troubleshoot, or review a Hume Empathic Voice Interface (EVI) realtime voice agent. Manages live, streaming audio interactions (speech-to-speech), including voice selection, prompts, tool use, interruption, turn-taking, transcripts, and privacy/safety considerations. Distinguishes between EVI versions (3 vs 4-mini) and when to apply each, and describes integration via WebSocket and SDKs.
How it works
- EVI is a realtime speech-to-speech interface. The Chat WebSocket accepts streamed audio_input, session_settings, user_input, assistant_input, and tool responses; audio input should be streamed in small chunks (roughly 20 ms audio buffers, or 100 ms for web apps).
- Use WebSocket-based browser/app integration with a React SDK or lower-level TS/Python SDKs. Audio capture can be in-browser for hosted web apps; Python is for CLIs/desktop where microphone access is local.
- EVI can use native Hume voices or account-private Custom Voices; voice is configurable per session or per config.
- Tools/function calling are supported via supplemental LLMs (not universal for all EVI configs); tool use follows OpenAI-style function calling conventions with supported LLMs.
- Expression measurements come from audio messages; transcripts/audio artifacts depend on data retention settings.
- Data retention governs chat history and audio reconstruction availability.
When to use it
Use this skill when designing, implementing, troubleshooting, or reviewing a Hume EVI realtime voice agent. Do not use it for offline-only Hume Text-to-Speech/Octave work unless configuring EVI voices or comparing EVI to TTS.
What it can touch
- WebSocket chat endpoint and streaming audio inputs/outputs.
- Configs and session settings for per-call context.
- Voice selection (EVI 3 vs EVI 4-mini) and persistent config options.
- Tools/function calling via supported supplemental LLMs.
- Audio capture, microphone handling, and playback pathways (client-side preferred).
- Data retention settings for chat history and audio reconstruction.
Caveats
- EVI is live and streamed; whole audio files are not supported as input for realtime conversations.
- Tool calling requires a supplemental or custom LLM configured for tool use.
- HIPAA readiness requires explicit BAAs; PHI handling depends on agreements.
- Data retention governs availability of chat history and audio reconstruction.
# Hume EVI production guide Use this skill when the task is to design, implement, troubleshoot, or review a Hume Empathic Voice Interface (EVI) realtime voice agent. Do not use it for offline-only Hume Text-to-Speech/Octave work unless the user is explicitly configuring EVI voices or comparing EVI to TTS. Hume's EVI is a realtime speech-to-speech agent interface. It streams user audio, measures ex
What does the hume-evi skill do?
Build production realtime voice agents with Hume's Empathic Voice Interface (EVI): speech-to-speech sessions, empathic/prosodic response design, EVI 3 versus EVI 4-mini selection, WebSocket and SDK integration, tools/function calling, interruption/barge-in, turn-taking, transcripts/audio artifacts, pricing/limits, privacy, safety, consent, and QA.
How do I install it?
Run `npx skills add calesthio/generative-media-skills --skill hume-evi --agent claude-code` — it drops the skill into your project so the agent can pick it up. Swap the --agent value for codex, cursor or copilot if you use one of those.
Where does this skill come from?
From calesthio/generative-media-skills, a repository with 112 stars. We read it straight from the repository tree rather than a submitted listing, so what you see here is what is actually published.
Is a popular skill a good skill?
Not necessarily. Stars measure attention, not adoption — a repository can trend for a week and be abandoned. That is why we show the weekly change from our own snapshots next to the total, instead of a single flattering number.