Spaces:

sarulab-speech
/

MSR-UTMOS

Running on Zero

Wataru commited on Jul 19

Commit

5960d7e

1 Parent(s): 48fe9ce

update reamde

Files changed (1) hide show

README.md CHANGED Viewed

@@ -8,7 +8,10 @@ sdk_version: 5.38.0
 app_file: app.py
 pinned: false
 license: mit
 short_description: Multiple sampling rate MOS prediction with SFI conv
 ---
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

 app_file: app.py
 pinned: false
 license: mit
+python_version: 3.12
 short_description: Multiple sampling rate MOS prediction with SFI conv
 ---
+In this space we showcase the prediction demo of "MSR-UTMOS".
+This demo uses sfi-w2v2 with pretraining on BVCC dataset.
+Please note that the prediction only uses fold0 of model. For the challenge submission, we used prediction results from all 7 folds.