MiniMax Speech 2.6 Turbo

MiniMax Speech 2.6 Turbo · MiniMax — Text to audio · 15 voices · MP3/WAV/FLAC. Call it through one queue API and pay for the options you pick.

modellane/minimax-speech-2.6-turbo · Text to audio · MiniMax

Text to audio model: 15 voices, MP3/WAV/FLAC.

Text text — required, at most 10000 characters The text fed into the generation.
Voice voice — optional, options: English expressive narrator, English radiant girl, English magnetic voiced man, English compelling lady1, English Aussie Bloke, English captivating female1, English Upbeat Woman, English Trustworth Man, English CalmWoman, English UpsetGirl, English Gentle voiced man, English Whispering girl, English Diligent Man, English Graceful Lady, English ReservedYoungMan The voice profile to use.
Speed speed — optional, default: 1, range: 0.5–2
Volume vol — optional, default: 1, range: 0–10
Pitch pitch — optional, default: 0, range: -12–12
Emotion emotion — optional, options: Auto, Happy, Sad, Angry, Fearful, Disgusted, Surprised, Calm, Fluent, Neutral
Language boost language_boost — optional, options: Auto, English, Chinese, Chinese,Yue, Japanese, Korean, Spanish, French, Portuguese, German, Italian, Arabic, Russian, Indonesian, Thai, Vietnamese
File format format — optional, options: MP3, WAV, FLAC The file format of the output.
Sample rate sample_rate — optional, default: 32000, options: 8000, 16000, 22050, 24000, 32000, 44100 The sample rate of the audio output; a higher value makes the file larger.
Bitrate bitrate — optional, default: 128000, options: 32000, 64000, 128000, 256000

Output and retention

The generated file and your input are kept for seven days and then deleted.

Submit to the queue: POST /v1/queue/modellane/minimax-speech-2.6-turbo

Status, result, cancellation and error codes work the same for every model; see the API docs.

Everything on this page is public; you need an account to run the model.