DAJA (MVDR,HMA,1000) (Overlapped Speech)
Reported on 2 benchmarks across 2 tasks · 1 paper · 2 SOTA
Note: results are matched by exact model name. Different papers may use the same name for different model variants.
Audio2 results
- WER (%)· 2022-07-15SOTA62.36best: 52.1 (ReVISE (bf))
- SDR· 2022-07-15SOTA-4.76