DrishtiSharma commited on
Commit
e55a21a
1 Parent(s): 48e4298

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +32 -9
README.md CHANGED
@@ -28,6 +28,33 @@ model-index:
28
  - name: Test CER
29
  type: cer
30
  value: 0.051648321634392154
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
31
  ---
32
 
33
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -40,19 +67,15 @@ It achieves the following results on the evaluation set:
40
  - Loss: 0.2756
41
  - Wer: 0.2279
42
 
43
- ## Model description
44
-
45
- More information needed
46
-
47
- ## Intended uses & limitations
48
 
49
- More information needed
50
 
51
- ## Training and evaluation data
52
 
53
- More information needed
54
 
55
- ## Training procedure
56
 
57
  ### Training hyperparameters
58
 
 
28
  - name: Test CER
29
  type: cer
30
  value: 0.051648321634392154
31
+ - name: Test WER (+LM)
32
+ type: wer
33
+ value: 0.13482652613087395
34
+ - name: Test CER (+LM)
35
+ type: cer
36
+ value: 0.038838663862562475
37
+ - task:
38
+ name: Automatic Speech Recognition
39
+ type: automatic-speech-recognition
40
+ dataset:
41
+ name: Robust Speech Event - Dev Data
42
+ type: speech-recognition-community-v2/dev_data
43
+ args: sl
44
+ metrics:
45
+ - name: Dev WER
46
+ type: wer
47
+ value: 0.5406156320830592
48
+ - name: Dev CER
49
+ type: cer
50
+ value: 0.22249723590310583
51
+ - name: Dev WER (+LM)
52
+ type: wer
53
+ value:
54
+ - name: Dev CER (+LM)
55
+ type: cer
56
+ value:
57
+
58
  ---
59
 
60
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 
67
  - Loss: 0.2756
68
  - Wer: 0.2279
69
 
70
+ ### Evaluation Commands
 
 
 
 
71
 
72
+ 1. To evaluate on mozilla-foundation/common_voice_8_0 with test split
73
 
74
+ python eval.py --model_id DrishtiSharma/wav2vec2-large-xls-r-300m-sl-with-LM-v1 --dataset mozilla-foundation/common_voice_8_0 --config sl --split test --log_outputs
75
 
76
+ 2. To evaluate on speech-recognition-community-v2/dev_data
77
 
78
+ python eval.py --model_id DrishtiSharma/wav2vec2-large-xls-r-300m-sl-with-LM-v1 --dataset speech-recognition-community-v2/dev_data --config sl --split validation --chunk_length_s 10 --stride_length_s 1
79
 
80
  ### Training hyperparameters
81