Search
Xiaomi Open-Sources MiDashengLM-Spatial, a Large Model for Spatial Audio Understanding
3 day ago /
Read about 0 minute
Author：小编
On October 15, 2026, Xiaomi's technical team released and open-sourced MiDashengLM-Spatial, a new model in the Dasheng series. Built upon MiDashengLM-7B, this model not only understands audio content but also perceives the source and spatial distribution of sounds. As the industry's first open-source end-to-end large model for spatial audio understanding, it will make its model weights and inference code publicly available, allowing developers, industry partners, and academic institutions worldwide to freely use it for commercial product integration or academic research. The model will be applied in scenarios such as Xiaomi's smart home and in-car cabin systems, enhancing the interactive experience across the 'Human-Car-Home Ecosystem.'
-
C114 Communication Network -
Communication Home
7 X 24 Track global technological trends
