Skip to content

Text to Speech

Module · Stable

Synthesize speech audio from text via OpenAI TTS. Returns base64-encoded MP3.

Calls the OpenAI TTS API (requires OPENAI_API_KEY). Voices: alloy, echo, fable, onyx, nova, shimmer. Models: tts-1 (fast) or tts-1-hd (higher quality). Returns base64-encoded MP3 audio suitable for WhatsApp voice replies or file storage.

Use when you have a text string — produced by Format, Ask AI, or extracted from context — that should reach a recipient as spoken audio. The audio_base64 output feeds directly into Send WhatsApp (voice reply) or Document Store (save the file for later retrieval).

To go the other direction (audio → text), use Speech to Text instead. If you only need to deliver a written message, skip synthesis entirely and use Send WhatsApp or Send email — TTS adds latency and API cost for no benefit when the recipient reads rather than listens.

Configured per use: text, voice, model, speed.

  • audio_base64
  • mime_type
  • engine
  • model
  • error

Auto-generated from the skill registry (load_skills()). Do not edit by hand.