Actions8
- Speech Actions
- Audio Actions
- Voices Actions
- Account Actions
Speech → Generate
AI-generatedSummary
Generates speech from text using the LeanVox TTS API, supporting synchronous and asynchronous modes, and three models (Standard, Pro, Max).
Inputs
- text (required) — The text to convert to speech.
- model (required) — The TTS model to use: Standard (fast & lightweight), Pro (238 curated voices with cloning), or Max (instruction-based voice design).
- voice (required) — The voice ID to use for synthesis.
- language — Optional language code (e.g., en, es, fr).
- voiceInstructions — Custom voice instructions (only valid with model 'max').
Output shape
A single JSON object containing the audio URL (audio_url) and job details for the generated audio.
For the 'Generate' operation, the audio is returned synchronously. For 'Generate Async', the response includes the job ID; the audio URL becomes available after processing.
Examples
Example 1: Generate speech synchronously using a Standard model.
Set Model to 'Standard', provide a Text and Voice ID.
Example 2: Generate speech asynchronously with voice instructions.
Set Model to 'Max', provide Text, Voice ID, and Voice Instructions.