The authors extended MiDashengLM, an 8B-parameter large audio-language model (LALM) designed for monaural input, with a spatial encoder called Spatial-Dasheng.
Unified model blends general audio understanding with spatial awareness
MiDashengLM-Spatial extends a monaural large audio-language model with a dedicated spatial encoder while preserving its general capabilities.
Industry
Jinbo Hu · Hang Su · Lichun Fan · Heinrich Dinkel · Gang Li · Zhanchen Dai · +7 more
Xiaomi Inc.
Research Digest··2 min read
The authors present MiDashengLM-Spatial, the first open-source end-to-end audio-language model that jointly handles general audio understanding and spatial awareness.
Why this paper
From Xiaomi Inc.
In one line
MiDashengLM-Spatial unifies general audio understanding and spatial awareness in one open-source audio-language model.
What we could check
- ·No code link found
- ·No weights link found
- ·No dataset link found
- ·No compute details found
- ·No stated limitations found
- ·No benchmark numbers found
Observed from the paper text and links we have. Absence here means we did not find it, not that it does not exist.
§