MiMo Audio 7B Instruct
MiMo Audio 7B Instruct adds instruction tuning to the base audio model. It supports spoken dialogue, audio understanding and text-to-speech controlled by instructions.
This is broader than a dedicated transcriber
The model card evaluates spoken dialogue and instruction-driven speech synthesis as well as understanding. It supplies separate inference examples for the base and instruction-tuned models.MiMo-Audio-7B-Instruct official model card
Evaluate the conversational turn as a whole
Practical guidance. A voice assistant must preserve the meaning of the input, produce a useful answer and speak it clearly. Check those steps separately. An attractive voice can mask an incorrect interpretation of the user's request.
Specifications & access
| Detail | Value and conditions |
|---|---|
| Access | Downloadable |
| Model identifiers | MiMo-Audio-7B-Instruct |
| Post-training | The release introduces thinking mechanisms for both audio understanding and generation.MiMo-Audio-7B-Instruct official model card |
| Licence | MIT.MiMo-Audio-7B-Instruct official model card |
| Local package | The guide loads the instruction model with the separate MiMo audio tokenizer.MiMo-Audio-7B-Instruct official model card |
Related models
- MiMo Audio 7B Base
Xiaomi · MiMo Audio
MiMo Audio 7B Base is a downloadable audio-language model trained at large scale.
Launch date not confirmedResearch releaseDownloadable - MiMo V2.5 ASR
Xiaomi · MiMo V2.5 Speech
V2.5 ASR targets transcription across Mandarin, English, Chinese dialects, mixed-language speech and noisy recordings.
Launch date not confirmedAPI · Downloadable - 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