The model is available on the hugging face platform with demos and scales, distributed under the Apache 2.0 license, and already shows excellent results in the 22 test of multimodal models and whispering even in tasks other than speech recognition.
Midashenglm-7B combines sound control, analysis of surrounding sounds and music, supports 50+ languages and recognizes individual users without activation words.
It analyzes the emotions and spatial features of the sound equipped with semantic mapping, which makes it universal for real time use in household appliances, electric vehicles and education systems.
The discovery of the code emphasizes the Xiaomi strategy to develop the “Man-Baskin” ecosystem and to compete with the Western AI giants through openness.
Source: Ferra

I am a professional journalist and content creator with extensive experience writing for news websites. I currently work as an author at Gadget Onus, where I specialize in covering hot news topics. My written pieces have been published on some of the biggest media outlets around the world, including The Guardian and BBC News.