Hugging Face's logo

Ilyes
/
wav2vec2-large-xlsr-53-french

Automatic Speech Recognition Transformers PyTorch
French wav2vec2 audio speech xlsr-fine-tuning-week Eval Results Inference Endpoints
Model card Files Community
3
Ilyes commited on
Commit
3e73b58
β€’
1 Parent(s): 252499b

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +2 -2
README.md CHANGED
@@ -45,7 +45,7 @@ ds = load_dataset("common_voice", "fr", split="test", cache_dir="./data/fr")
45
 
46
 
47
 
48
- chars_to_ignore_regex = '[\\,\\?\\.\\!\\-\\;\\:\\"\\β€œ\\%\\β€˜\\”\\οΏ½\\β€˜\\’\\’\\’\\β€˜\\…\\Β·\\!\\Ηƒ\\?\\Β«\\β€Ή\\Β»\\β€Ίβ€œ\\”\\\\ΚΏ\\ΚΎ\\β€ž\\∞\\\\|\\.\\,\\;\\:\\*\\β€”\\–\\─\\―\\_\\/\\:\\ː\\;\\,\\=\\Β«\\Β»\\β†’]'
49
  def map_to_array(batch):
50
  speech, _ = torchaudio.load(batch["path"])
51
  batch["speech"] = resampler.forward(speech.squeeze(0)).numpy()
@@ -74,7 +74,7 @@ print(wer.compute(predictions=result["predicted"], references=result["target"]))
74
 
75
  ## Training
76
 
77
- 6% of the Common Voice `train`, `validation` datasets were used for training.
78
 
79
  ## Testing
80
 
45
 
46
 
47
 
48
+ chars_to_ignore_regex = '[\\\\,\\\\?\\\\.\\\\!\\\\-\\\\;\\\\:\\\\"\\\\β€œ\\\\%\\\\β€˜\\\\”\\\\οΏ½\\\\β€˜\\\\’\\\\’\\\\’\\\\β€˜\\\\…\\\\Β·\\\\!\\\\Ηƒ\\\\?\\\\Β«\\\\β€Ή\\\\Β»\\\\β€Ίβ€œ\\\\”\\\\\\\\ΚΏ\\\\ΚΎ\\\\β€ž\\\\∞\\\\\\\\|\\\\.\\\\,\\\\;\\\\:\\\\*\\\\β€”\\\\–\\\\─\\\\―\\\\_\\\\/\\\\:\\\\ː\\\\;\\\\,\\\\=\\\\Β«\\\\Β»\\\\β†’]'
49
  def map_to_array(batch):
50
  speech, _ = torchaudio.load(batch["path"])
51
  batch["speech"] = resampler.forward(speech.squeeze(0)).numpy()
74
 
75
  ## Training
76
 
77
+ 6% of the Common Voice `train`, `validation` datasets (20K files) were used for training.
78
 
79
  ## Testing
80