MiMo V2.5 TTS VoiceDesign
VoiceDesign creates a new voice from a written description. It does not need a preset voice or an uploaded sample.
Describe the sound rather than studio processing
The guide accepts Chinese or English descriptions. It recommends concise, non-conflicting voice traits and warns against treating effects such as reverb or equalization as voice characteristics.MiMo V2.5 speech synthesis guide
Keep identity and delivery separate
Practical guidance. Describe a stable voice identity first, then adjust pace or emotion for the script. Compare several samples before choosing a voice. A description that works for a calm narration may need different delivery instructions for an urgent announcement.
Specifications & access
| Detail | Value and conditions |
|---|---|
| Access | API |
| Model identifiers | mimo-v2.5-tts-voicedesign |
| Required direction | The user message must describe the voice. The assistant message contains the text to speak.MiMo speech synthesis API |
| Limits | This model does not support the singing mode, preset voices or voice cloning.MiMo V2.5 speech synthesis guide |
Related models
- MiMo V2.5 TTS
Xiaomi · MiMo V2.5 Speech
The standard V2.5 TTS model turns text into speech using Xiaomi's built-in voices.
Launch date not confirmedAPI - MiMo V2.5 TTS VoiceClone
Xiaomi · MiMo V2.5 Speech
VoiceClone generates speech using a voice sampled from audio.
Launch date not confirmedAPI