Hugging Face's logo

CuongLD
/
wav2vec2-large-xlsr-vietnamese

Automatic Speech Recognition
Transformers
PyTorch
JAX
Vietnamese
wav2vec2
audio
speech
xlsr-fine-tuning-week
Eval Results (legacy)
Model card Files
xet
 Community
1
CuongLD commited on
Commit
d6dacff
·
1 Parent(s): fb52b9b

modify readme

Browse files
Files changed (1) hide show
  1. README.md +1 -1
README.md CHANGED
@@ -86,7 +86,7 @@ processor = Wav2Vec2Processor.from_pretrained("CuongLD/wav2vec2-large-xlsr-vietn
86
  model = Wav2Vec2ForCTC.from_pretrained("CuongLD/wav2vec2-large-xlsr-vietnamese")
87
  model.to("cuda")
88
 
89
- chars_to_ignore_regex = '[\\\\\\\\,\\\\\\\\?\\\\\\\\.\\\\\\\\!\\\\\\\\-\\\\\\\\;\\\\\\\\:\\\\\\\\"\\\\\\\\“]'
90
  resampler = torchaudio.transforms.Resample(48_000, 16_000)
91
 
92
  # Preprocessing the datasets.
 
86
  model = Wav2Vec2ForCTC.from_pretrained("CuongLD/wav2vec2-large-xlsr-vietnamese")
87
  model.to("cuda")
88
 
89
+ chars_to_ignore_regex = '[\,\?\.\!\-\;\:\"\“]'
90
  resampler = torchaudio.transforms.Resample(48_000, 16_000)
91
 
92
  # Preprocessing the datasets.