bayartsogt commited on
Commit
b7ec1f5
1 Parent(s): 0cef473

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +4 -4
README.md CHANGED
@@ -11,7 +11,7 @@ tags:
11
  - xlsr-fine-tuning-week
12
  license: apache-2.0
13
  model-index:
14
- - name: wav2vec2-large-xlsr-mongolian-v1
15
  results:
16
  - task:
17
  name: Speech Recognition
@@ -23,7 +23,7 @@ model-index:
23
  metrics:
24
  - name: Test WER
25
  type: wer
26
- value: 35.42
27
  ---
28
 
29
  # Wav2Vec2-Large-XLSR-53-Mongolian-v1
@@ -72,7 +72,7 @@ def evaluate(batch):
72
  with torch.no_grad():
73
  logits = model(inputs.input_values.to("cuda"), attention_mask=inputs.attention_mask.to("cuda")).logits
74
 
75
- pred_ids = torch.argmax(logits, dim=-
76
  batch["pred_strings"] = processor.batch_decode(pred_ids)
77
  return batch
78
 
@@ -81,7 +81,7 @@ result = test_dataset.map(evaluate, batched=True, batch_size=8)
81
  print("WER: {:2f}".format(100 * wer.compute(predictions=result["pred_strings"], references=result["sentence"])))
82
  ```
83
 
84
- **Test Result**: 35.42 %
85
 
86
 
87
  ## Training
 
11
  - xlsr-fine-tuning-week
12
  license: apache-2.0
13
  model-index:
14
+ - name: XLSR Wav2Vec2 Mongolian V1 by Bayartsogt
15
  results:
16
  - task:
17
  name: Speech Recognition
 
23
  metrics:
24
  - name: Test WER
25
  type: wer
26
+ value: 35.33
27
  ---
28
 
29
  # Wav2Vec2-Large-XLSR-53-Mongolian-v1
 
72
  with torch.no_grad():
73
  logits = model(inputs.input_values.to("cuda"), attention_mask=inputs.attention_mask.to("cuda")).logits
74
 
75
+ pred_ids = torch.argmax(logits, dim=-1)
76
  batch["pred_strings"] = processor.batch_decode(pred_ids)
77
  return batch
78
 
 
81
  print("WER: {:2f}".format(100 * wer.compute(predictions=result["pred_strings"], references=result["sentence"])))
82
  ```
83
 
84
+ **Test Result**: 35.33 %
85
 
86
 
87
  ## Training