Pricing is temporarily unavailable — capabilities and docs are still fully viewable.
Billing details
Description
Xiaomi MiMo v2.5 TTS — expressive text-to-speech with eight built-in voices, four Chinese and four English (Mia, Chloe, Milo, Dean), covering Chinese and English plus Northeastern, Sichuanese, Henan, and Cantonese dialect styles. Delivery is directed two ways: natural-language style instructions, and inline audio tags for basic and compound emotions, breathing, sighing, and pacing, including a singing mode unique to this model in the MiMo TTS line. Returns 24 kHz mono wav or pcm16 and supports low-latency streaming, which requires pcm16. 8K context and 8K maximum output.
Capabilities
- Tasks
- text-to-speech
API endpoints
POST /v1/audio/speech(audio-speech)