Automatic Speech Recognition
ESPnet
English
audio
wanchichen commited on
Commit
d6afe74
·
verified ·
1 Parent(s): 189f4ba

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +3 -0
README.md CHANGED
@@ -11,6 +11,9 @@ license: cc-by-4.0
11
 
12
  ## ESPnet2 ASR model
13
 
 
 
 
14
  ### Demo: How to use in ESPnet2
15
 
16
  Follow the [ESPnet installation instructions](https://espnet.github.io/espnet/installation.html)
 
11
 
12
  ## ESPnet2 ASR model
13
 
14
+ This is a simple baseline for the ML-SUPERB 2.0 Challenge. It is a self-supervised [MMS 1B](https://huggingface.co/facebook/mms-1b) model fine-tuned on [142 languages of ML-SUPERB](https://huggingface.co/datasets/ftshijt/mlsuperb_8th) using CTC loss.
15
+ The MMS model is frozen and used as a feature extractor for a small Transformer encoder during fine-tuning, which took approximately 1 day on a single GPU.
16
+
17
  ### Demo: How to use in ESPnet2
18
 
19
  Follow the [ESPnet installation instructions](https://espnet.github.io/espnet/installation.html)