Smallest AI icon

Smallest AI

Interact with Smallest AI APIs for voice synthesis and agents

Actions6

Speech (TTS) → Synthesize Speech

AI-generated

Summary

Convert input text to speech audio using the Smallest AI Lightning V3.1 model, with choice of built-in or custom voice, output audio format, and optional speech synthesis parameters.

Inputs

  • Model (required) — Select the text-to-speech model to use (currently only Lightning V3.1).
  • Text (required) — Provide the text string to convert into spoken audio.
  • Voice Source (required) — Choose between using a built-in voice or a custom/cloned voice ID for synthesis.
  • Voice (required) — If using a built-in voice, select one of the extensive predefined voices supporting multiple languages and accents.
  • Custom Voice ID (required) — If using a custom voice source, enter the cloned or custom voice ID to use.
  • Output Format (required) — Specify the audio format of the synthesized speech output (MP3, Mulaw, PCM, WAV).
  • Additional Options — Optionally configure language (auto, English, Hindi, Spanish, Tamil), sample rate (8000, 16000, 24000, 44100 Hz), and speech speed (0.5–2x) for the synthesized audio.

Output shape

a single item containing the synthesized speech as binary data and a JSON object confirming success and format

The output includes a binary property named 'data' containing the audio in the requested format, and a JSON object with 'success' as true and the audio format; this binary data can be used directly or saved as an audio file in downstream workflow nodes.

Links

Discussion