DEC 23, 2024
December 23, 2024
From December 2024, kept as a record. Speech has changed a great deal since, and is no longer free. For what it does today see POST speech/create and How to Generate Text-to-Speech Audio with the MiniMax API.
Added support for TTS (text-to-speech) to MiniMax API v1.
Both MP3 export POST speech/create-mp3 and live streaming POST speech/create-stream are available.
- Up to 20 parallel TTS jobs per account are supported.
- Average time to response for live streaming is 3 seconds.
- Currently, this service is offered free of charge.
Over 300 pre-built voices provided GET speech/voices supporting the following:
- Languages: English, Chinese (Mandarin), Spanish, French, Russian, Portuguese, Indonesian, German, Japanese, Korean, Italian, Cantonese
- Emotions: happy, sad, angry, fearful, disgusted, surprised, neutral
- Accents: US (General), English, Indian
- Ages: Young Adult, Adult, Middle-Aged, Senior
- Genders: Male, Female
We made sure you can fully explore all features using the documentation pages linked above. See the Try It section for more details.
Examples below were created using the MiniMax API endpoint POST speech/create-mp3.
- May the Force be with you reference
- There’s no place like home reference
- Yippee-Ki-Yay reference
- Dr. Evil: Sharks with laser beams attached to their heads reference