Actions10
音声合成（簡略）
AI-generatedSummary
Synthesize speech audio from input text using a specified speaker style and optional speech parameter adjustments, returning the audio as a binary WAV file.
Inputs
- text (required) — The text to be synthesized into speech.
- speakerId (required) — The ID of the speaker style to use for synthesis, as obtained from the speaker list.
- audioParams — Optional collection of speech synthesis parameters to adjust speed, pitch, intonation, volume, silence padding, and tempo dynamics; if not provided, API defaults are used.
- binaryPropertyName — Name of the binary property under which the resulting WAV audio file is stored; defaults to 'data'.
- outputStereo — Boolean flag to output stereo (2 channel) audio; defaults to false (mono).
Output shape
one item with JSON success status and input parameters, plus a binary property containing the synthesized WAV audio
The operation returns an object with a JSON body indicating success, the original text, and the speaker ID used, alongside a binary attachment containing the audio file. The binary property name can be customized and is accessed in subsequent nodes via $binary.[binaryPropertyName]. The synthesis internally performs AudioQuery retrieval followed by synthesis with applied parameters.