VoiceThere
Sign up

Google Cloud speech on VoiceThere

Google Chirp streaming STT and Neural2 or Chirp3-HD TTS on VoiceThere cloud with service account or API key secrets.

Overview of all vendors: STT & TTS vendors.

What it is

Google Cloud Speech-to-Text and Text-to-Speech APIs back this vendor. Chirp models use V2 streaming recognition when a recognizer path is configured; Neural2 and Chirp3-HD cover TTS.

Home: Chirp 3 speech-to-text, Streaming text-to-speech

Set up on VoiceThere

  • Dashboard: Project Settings → Voice (STT / TTS). Pick google 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_GOOGLE_APPLICATION_CREDENTIALS: Service account credentials. Service-account JSON content or absolute path to the credentials file on the runner.
  • AGENT_GOOGLE_API_KEY: Google API key. API key for Google Speech endpoints when not using ADC alone.

CLI (after secrets exist):

voicethere projects voice set --project YOUR_PROJECT_ID \
  --stt-provider google --tts-provider google \
  --stt-model chirp_3 --tts-voice en-US-Chirp3-HD-Charon

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: latest_long, chirp_3

TTS voice examples: en-US-Neural2-A, af-ZA-Standard-A

Store service account JSON in AGENT_GOOGLE_APPLICATION_CREDENTIALS and/or set AGENT_GOOGLE_API_KEY when using API key auth. Chirp streaming STT may require project, location, and recognizer configuration on the runner when your Google setup uses named recognizers.

What the implementation does

  • chirp_3, chirp_2, telephony: V2 StreamingRecognize with partials when configured.
  • latest_long: batch REST recognition, no live partials.
  • Chirp3-HD voices stream TTS audio; Neural2 voices use batch REST synthesis.

Limitations

  • Billed by Google Cloud.
  • Streaming Chirp STT needs a valid recognizer resource path in your Google project; misconfigured ADC or recognizer paths fail at runtime.

Official docs

← All documentation