anuragshas commited on
Commit
dfc5f1c
β€’
1 Parent(s): 6fabf17

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +2 -2
README.md CHANGED
@@ -25,7 +25,7 @@ model-index:
25
  type: wer
26
  value: 55.68
27
  ---
28
- # Wav2Vec2-Large-XLSR-53-Odia
29
  Fine-tuned [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on Dhivehi using the [Common Voice](https://huggingface.co/datasets/common_voice).
30
  When using this model, make sure that your speech input is sampled at 16kHz.
31
  ## Usage
@@ -66,7 +66,7 @@ wer = load_metric("wer")
66
  processor = Wav2Vec2Processor.from_pretrained("anuragshas/wav2vec2-large-xlsr-53-dv")
67
  model = Wav2Vec2ForCTC.from_pretrained("anuragshas/wav2vec2-large-xlsr-53-dv")
68
  model.to("cuda")
69
- chars_to_ignore_regex = '[\,\?\.\!\-\;\:\"\β€œ\%\β€˜\”\،\؟\–\'\’]'
70
  resampler = torchaudio.transforms.Resample(48_000, 16_000)
71
  # Preprocessing the datasets.
72
  # We need to read the aduio files as arrays
 
25
  type: wer
26
  value: 55.68
27
  ---
28
+ # Wav2Vec2-Large-XLSR-53-Dhivehi
29
  Fine-tuned [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on Dhivehi using the [Common Voice](https://huggingface.co/datasets/common_voice).
30
  When using this model, make sure that your speech input is sampled at 16kHz.
31
  ## Usage
 
66
  processor = Wav2Vec2Processor.from_pretrained("anuragshas/wav2vec2-large-xlsr-53-dv")
67
  model = Wav2Vec2ForCTC.from_pretrained("anuragshas/wav2vec2-large-xlsr-53-dv")
68
  model.to("cuda")
69
+ chars_to_ignore_regex = '[\,\?\.\!\-\;\:\"\β€œ\%\β€˜\”\،\.\؟\–\'\’]'
70
  resampler = torchaudio.transforms.Resample(48_000, 16_000)
71
  # Preprocessing the datasets.
72
  # We need to read the aduio files as arrays