#
dnsmos
Here are 3 public repositories matching this topic...
audio filtering pipeline for large-scale Indic speech datasets
-
Updated
Apr 19, 2026 - Python
Autoregressive speech prediction with EnCodec and FACodec token language models: 2s context -> 1s future speech, with STOI/PESQ/DNSMOS analysis of where those metrics disagree.
pytorch speech-synthesis transformer language-model speech-processing pesq stoi encodec dnsmos neural-audio-codec facodec speech-quality-assessment
-
Updated
Aug 6, 2026 - Python
Add this topic to your repo
To associate your repository with the dnsmos topic, visit your repo's landing page and select "manage topics."