Hume Evi Sessions

Agent opens wss://api.hume.ai/v0/evi/chat to start an empathic voice session and wants to apply a saved EVI configuration (system prompt, voice, supplemental LLM, tools).

Reference the saved configuration by passing config_id as a query parameter on the chat WebSocket URL. Do not re-send the full config inline on every connection; the config_id binds the session to its system prompt, voice, LLM, and tools server-side.

Server streams user_message (transcribed user speech, with prosody scores) and assistant_message events. The UI renders both into one undifferentiated transcript.

Route user_message and assistant_message to their respective roles in the transcript; attach any prosody scores carried on the message as auxiliary signals, not as the spoken text. Preserve event order for an accurate conversation log.

Client captures mic audio as 16-bit PCM at 16000 Hz mono but the EVI session defaults assume a different encoding, producing garbled ASR.

Send a session_settings message immediately after connect declaring the audio encoding, sample_rate, and channels that match the client capture (e.g. linear16 / 16000 / 1) before streaming audio_input. Re-declare on reconnect.