Maia RouterInstall

Summary

Generate audio from text using a selected TTS model and voice, returning the audio as binary data plus metadata.

Inputs

• Model (required) — ID of the TTS model to use
• Input Text (required) — The text to generate audio for. Maximum length is 4096 characters.
• Voice (required) — Voice to use for the generated audio.
• Additional Fields — Optional settings for the audio generation.

Output shape

json + binary

JSON includes success, model, voice, format, input, and size (bytes). Binary contains the generated audio file named speech. with MIME type audio/.

Examples

Example 1: Convert a short notification to speech

Model: openai/gpt-4o-mini-tts, Input Text: 'Hello world', Voice: alloy, Additional Fields: { response_format: 'mp3', speed: 1 }