@soniox/client

Official Soniox React SDK for React, React Native and Next.js

Full React SDK Documentation

Installation

npm install @soniox/react

Realtime TTS timestamps

In WebSocket mode, useTts can return character-level audio timestamps. Set return_timestamps: true on the hook config; the alignment for each frame is delivered as the second argument of onAudio (it is undefined for audio-only frames). Timestamps are WebSocket-only — REST mode never provides them — and map to the model's preprocessed text, not your raw input.

const { speak } = useTts({
  config,
  voice: 'Adrian',
  return_timestamps: true,
  onAudio: (chunk, timestamps) => {
    play(chunk);
    if (timestamps) {
      // timestamps.characters / character_start_times_seconds / character_end_times_seconds
    }
  },
});

speak('Hello world');

For the full documentation please go to our docs: Full React SDK Documentation