Voice (TTS / STT)
tota supports multiple TTS and STT providers. Run tota setup voice to configure interactively.
#TTS providers
| Env Var | Description | Default |
|---|---|---|
OPENAI_API_KEY | Used for OpenAI TTS-1 | — |
ELEVENLABS_API_KEY | ElevenLabs API key | — |
ELEVENLABS_VOICE_ID | ElevenLabs voice ID to use | — |
GOOGLE_TTS_API_KEY | Google Cloud TTS API key | — |
#STT providers
| Env Var | Description | Default |
|---|---|---|
OPENAI_API_KEY | Used for OpenAI Whisper-1 | — |
GROQ_API_KEY | Groq Whisper-large-v3 (faster) | — |
Or configure fully in ~/.tota/tota.yaml:
voice:
ttsProvider: openai # openai | elevenlabs | google
sttProvider: openai # openai | groq
defaultVoice: nova # alloy | echo | fable | onyx | nova | shimmer
elevenLabsApiKey: sk-...
elevenLabsVoiceId: VOICE_ID
googleTtsApiKey: AIza...
groqApiKey: gsk_...
The provider param on text_to_speech and transcribe_audio tools overrides the configured default per call.
When any STT key is present, Telegram voice messages are automatically transcribed before being passed to the agent.
Run tota setup voice to configure providers via an interactive wizard.