VoiceThere
Sign up

Deepgram speech on VoiceThere

Deepgram Listen STT over WebSocket and Aura or Flux TTS on VoiceThere cloud with AGENT_DEEPGRAM_API_KEY.

Overview of all vendors: STT & TTS vendors.

What it is

Deepgram Listen handles streaming STT; Aura and Flux models handle TTS. One API key covers both legs when you select Deepgram for STT and/or TTS.

Home: Deepgram live streaming STT, Deepgram Speak TTS

Set up on VoiceThere

  • Dashboard: Project Settings → Voice (STT / TTS). Pick deepgram on the leg this vendor supports. Add project secrets from the list below, then Deploy to cloud.
  • Secrets: Store keys as AGENT_* project secrets (see Agent environment & secrets).
  • API: Patch stt_provider / tts_provider on the project before deploy. See Control plane API.

Project secrets for this vendor

  • AGENT_DEEPGRAM_API_KEY: Deepgram API key. Streaming listen STT and speak TTS on the runner.

CLI (after secrets exist):

voicethere projects voice set --project YOUR_PROJECT_ID \
  --stt-provider deepgram --tts-provider deepgram \
  --stt-model nova-2 --tts-model aura-asteria-en

Models and voices you can pick

Project Voice settings in the dashboard lists every model and voice id VoiceThere accepts for this vendor. The same allowlists ship in the npm package @node-webrtc-rust/voice-catalog (pinned on your platform release).

STT examples: nova-2, nova-3

TTS model examples: aura-asteria-en, aura-luna-en

What the implementation does

STT opens a Listen WebSocket with nova-2 or nova-3 only. Flux is not used on the listen path. TTS calls Aura models on /v1/speak or Flux models on /v2/speak based on the model id prefix. Partials are available on the Listen stream.

Limitations

  • Flux models are for TTS, not Listen STT.
  • Billed by Deepgram.

Official docs

← All documentation