Volcano TTSInstall

Summary

Convert input text (up to 2000 characters) into synthesized speech audio using Volcano Engine TTS with selectable voice and audio parameters.

Inputs

• Text (required) — The text to convert to speech, maximum 2000 characters.
• Voice (required) — Select the voice for speech synthesis from provided Chinese and English options.
• Additional Options — Customize audio output and synthesis including connection mode (binary, websocket, HTTP), audio format (mp3, wav, pcm, ogg), sample rate (8000, 16000, 24000, 48000 Hz), speech speed (0.5 to 2.0), volume (0-100), pitch (-12 to +12 semitones), emotion (neutral, happy, sad, angry, fearful, surprised), enable SSML parsing, output as binary or base64, and binary property name if applicable.

Output shape

A list of items, each containing JSON metadata and synthesized audio either as a binary property or base64 string, depending on the output format chosen.

Each output item includes a json object with success status, chosen voice, audio format, and length of input text. If outputAsBinary is true (default), the audio data is in a binary property named by binaryPropertyName (default 'audio') with appropriate MIME type; otherwise, audio is base64-encoded in JSON. Errors in processing individual items appear in JSON with an error message if 'continue on fail' is enabled.

Examples

Example 1: Convert text to speech with default voice and output as binary MP3.

Text set, voice defaulted to 通用女声 (中文), outputAsBinary true, format mp3.

Example 2: Synthesize speech with English male voice, WAV format, and increased speed.

Voice set to Male Voice (English), format set to wav, speed set to >1.0.