Srulikbdd
/

Wav2Vec2-large-xlsr-welsh

Automatic Speech Recognition

xlsr-fine-tuning-week

Inference Endpoints

Model card Files Files and versions Community

Srulikbdd commited on Mar 26, 2021

Commit

7124326

·

1 Parent(s): 1a0faec

Update README.md

Files changed (1) hide show

README.md +2 -2

README.md CHANGED Viewed

@@ -19,7 +19,7 @@ model-index:
     metrics:
        - name: Test WER
          type: wer
-         value: 20.93
 ---
@@ -80,7 +80,7 @@ model = Wav2Vec2ForCTC.from_pretrained("Srulikbdd/Wav2vec2-large-xlsr-welsh")
 model.to("cuda")
-chars_to_ignore_regex = '[\,\?\.\!\-\u2013\u2014\;\:\"\\%\\\]'
 resampler = torchaudio.transforms.Resample(48_000, 16_000)

     metrics:
        - name: Test WER
          type: wer
+         value: 29.4
 ---
 model.to("cuda")
+chars_to_ignore_regex = '[\\,\\?\\.\\!\\-\\u2013\\u2014\\;\\:\\"\\\\%\\\\\\]'
 resampler = torchaudio.transforms.Resample(48_000, 16_000)