Text to speech: how to choose a model
Text-to-speech converts a script into playable audio. Language, voice, speed, and format affect the integration.
Before integrating
- 01Choose the language and a voice you have permission to use.
- 02Check voice_id, speed range, and text limits.
- 03Determine whether the response is a file URL, binary data, or encoded audio.
- 04Check synchronous, streaming, or asynchronous flow and how results are saved.
Verify voice names, permissions, and API regions in your own account.
Example request: “Convert an article into natural narration with adjustable speaking speed”Find models for this scenario →
Related integrations
Browse the catalog →
Configuration under reviewMiniMax speech-2.8-hd
Convert text into narration using supported provider voices, including Chinese content.
Configuration under reviewMiniMax speech-01-hd
Convert text into narration using supported provider voices, including Chinese content.
Configuration under reviewMiniMax speech-01-turbo
Convert text into narration using supported provider voices, including Chinese content.
Configuration under reviewMiniMax speech-02-hd
Convert text into narration using supported provider voices, including Chinese content.
Configuration under reviewMiniMax speech-02-turbo
Convert text into narration using supported provider voices, including Chinese content.
Configuration under reviewMiniMax speech-2.6-hd
Convert text into narration using supported provider voices, including Chinese content.
Configuration under reviewMiniMax speech-2.6-turbo
Convert text into narration using supported provider voices, including Chinese content.
Configuration under reviewMiniMax speech-2.8-turbo
Convert text into narration using supported provider voices, including Chinese content.