MiniMax Speech·by MiniMaxAudio
MiniMax Speech 2.6 Turbo
MiniMax Speech 2.6 Turbo — low-latency speech synthesis with voice, speed and emotion control. Write the text to speak as the prompt (max 10,000 characters).
Open in workspace →
minimax/speech-2.6-turboParameters
| Parameter | Type | Default | Range or options |
|---|---|---|---|
Voice voiceMiniMax system voice ID. | select | English_expressive_narrator | 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 |
Bitrate bitrateOutput audio bitrate. | select | 128000 | 32000, 64000, 128000, 256000 |
Emotion emotionOptional emotion control. | select | auto | auto, happy, sad, angry, fearful, disgusted, surprised, calm, fluent, neutral |
Format formatOutput audio format. | select | mp3 | mp3, wav, flac |
Language language_boostOptional language hint. Use auto for automatic detection. | select | auto | auto, English, Chinese, Chinese,Yue, Japanese, Korean, Spanish, French, Portuguese, German, Italian, Arabic, Russian, Indonesian, Thai, Vietnamese |
Pitch pitchSpeech pitch adjustment. | slider | 0 | -12 – 12 |
Sample rate sample_rateOutput sample rate. | select | 32000 | 8000, 16000, 22050, 24000, 32000, 44100 |
Speed speedSpeech speed multiplier. | slider | 1 | 0.5 – 2 |
Volume volSpeech volume. | slider | 1 | 0 – 10 |
Inputs
No reference files: this model works from the prompt alone.