lucio commited on
Commit
3d92921
1 Parent(s): 1edb526

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +15 -15
README.md CHANGED
@@ -51,15 +51,15 @@ resampler = torchaudio.transforms.Resample(48_000, 16_000)
51
  # Preprocessing the datasets.
52
  # We need to read the audio files as arrays
53
  def speech_file_to_array_fn(batch):
54
- \tspeech_array, sampling_rate = torchaudio.load(batch["path"])
55
- \tbatch["speech"] = resampler(speech_array).squeeze().numpy()
56
- \treturn batch
57
 
58
  test_dataset = test_dataset.map(speech_file_to_array_fn)
59
  inputs = processor(test_dataset["speech"][:2], sampling_rate=16_000, return_tensors="pt", padding=True)
60
 
61
  with torch.no_grad():
62
- \tlogits = model(inputs.input_values, attention_mask=inputs.attention_mask).logits
63
 
64
  predicted_ids = torch.argmax(logits, dim=-1)
65
 
@@ -87,30 +87,30 @@ processor = Wav2Vec2Processor.from_pretrained("lucio/wav2vec2-large-xlsr-luganda
87
  model = Wav2Vec2ForCTC.from_pretrained("lucio/wav2vec2-large-xlsr-luganda")
88
  model.to("cuda")
89
 
90
- chars_to_ignore_regex = '[\\,\\?\\.\\!\\-\\;\\:\\"\\“]'
91
  resampler = torchaudio.transforms.Resample(48_000, 16_000)
92
 
93
  # Preprocessing the datasets.
94
  # We need to read the audio files as arrays
95
  def speech_file_to_array_fn(batch):
96
- \tbatch["sentence"] = re.sub(chars_to_ignore_regex, '', batch["sentence"]).lower()
97
- \tspeech_array, sampling_rate = torchaudio.load(batch["path"])
98
- \tbatch["speech"] = resampler(speech_array).squeeze().numpy()
99
- \treturn batch
100
 
101
  test_dataset = test_dataset.map(speech_file_to_array_fn)
102
 
103
  # Preprocessing the datasets.
104
  # We need to read the audio files as arrays
105
  def evaluate(batch):
106
- \tinputs = processor(batch["speech"], sampling_rate=16_000, return_tensors="pt", padding=True)
107
 
108
- \twith torch.no_grad():
109
- \t\tlogits = model(inputs.input_values.to("cuda"), attention_mask=inputs.attention_mask.to("cuda")).logits
110
 
111
- \tpred_ids = torch.argmax(logits, dim=-1)
112
- \tbatch["pred_strings"] = processor.batch_decode(pred_ids)
113
- \treturn batch
114
 
115
  result = test_dataset.map(evaluate, batched=True, batch_size=8)
116
 
 
51
  # Preprocessing the datasets.
52
  # We need to read the audio files as arrays
53
  def speech_file_to_array_fn(batch):
54
+ speech_array, sampling_rate = torchaudio.load(batch["path"])
55
+ batch["speech"] = resampler(speech_array).squeeze().numpy()
56
+ return batch
57
 
58
  test_dataset = test_dataset.map(speech_file_to_array_fn)
59
  inputs = processor(test_dataset["speech"][:2], sampling_rate=16_000, return_tensors="pt", padding=True)
60
 
61
  with torch.no_grad():
62
+ logits = model(inputs.input_values, attention_mask=inputs.attention_mask).logits
63
 
64
  predicted_ids = torch.argmax(logits, dim=-1)
65
 
 
87
  model = Wav2Vec2ForCTC.from_pretrained("lucio/wav2vec2-large-xlsr-luganda")
88
  model.to("cuda")
89
 
90
+ chars_to_ignore_regex = '[\,\?\.\!\-\;\:\"\“]'
91
  resampler = torchaudio.transforms.Resample(48_000, 16_000)
92
 
93
  # Preprocessing the datasets.
94
  # We need to read the audio files as arrays
95
  def speech_file_to_array_fn(batch):
96
+ batch["sentence"] = re.sub(chars_to_ignore_regex, '', batch["sentence"]).lower()
97
+ speech_array, sampling_rate = torchaudio.load(batch["path"])
98
+ batch["speech"] = resampler(speech_array).squeeze().numpy()
99
+ return batch
100
 
101
  test_dataset = test_dataset.map(speech_file_to_array_fn)
102
 
103
  # Preprocessing the datasets.
104
  # We need to read the audio files as arrays
105
  def evaluate(batch):
106
+ inputs = processor(batch["speech"], sampling_rate=16_000, return_tensors="pt", padding=True)
107
 
108
+ with torch.no_grad():
109
+ logits = model(inputs.input_values.to("cuda"), attention_mask=inputs.attention_mask.to("cuda")).logits
110
 
111
+ pred_ids = torch.argmax(logits, dim=-1)
112
+ batch["pred_strings"] = processor.batch_decode(pred_ids)
113
+ return batch
114
 
115
  result = test_dataset.map(evaluate, batched=True, batch_size=8)
116