Hugging Face's logo

jonatasgrosman
/
wav2vec2-xls-r-1b-russian

Automatic Speech Recognition Transformers PyTorch
Russian wav2vec2 hf-asr-leaderboard mozilla-foundation/common_voice_8_0 robust-speech-event Eval Results Inference Endpoints
Model card Files Community
2
anton-l HF staff commited on
Commit
be8fa13
•
1 Parent(s): 7bbf0ec

Upload README.md

Browse files
Files changed (1) hide show
  1. README.md +38 -27
README.md CHANGED
@@ -12,27 +12,27 @@ datasets:
12
  model-index:
13
  - name: XLS-R Wav2Vec2 Russian by Jonatas Grosman
14
  results:
15
- - task:
16
- name: Automatic Speech Recognition
17
  type: automatic-speech-recognition
18
  dataset:
19
  name: Common Voice 8
20
  type: mozilla-foundation/common_voice_8_0
21
  args: ru
22
  metrics:
23
- - name: Test WER
24
- type: wer
25
- value: 9.82
26
- - name: Test CER
27
- type: cer
28
- value: 2.30
29
- - name: Test WER (+LM)
30
- type: wer
31
- value: 7.08
32
- - name: Test CER (+LM)
33
- type: cer
34
- value: 1.87
35
- - task:
36
  name: Automatic Speech Recognition
37
  type: automatic-speech-recognition
38
  dataset:
@@ -40,18 +40,29 @@ model-index:
40
  type: speech-recognition-community-v2/dev_data
41
  args: ru
42
  metrics:
43
- - name: Dev WER
44
- type: wer
45
- value: 23.96
46
- - name: Dev CER
47
- type: cer
48
- value: 8.88
49
- - name: Dev WER (+LM)
50
- type: wer
51
- value: 15.88
52
- - name: Dev CER (+LM)
53
- type: cer
54
- value: 7.42
 
 
 
 
 
 
 
 
 
 
 
55
  ---
56
 
57
  # XLS-R-1B-RUSSIAN
12
  model-index:
13
  - name: XLS-R Wav2Vec2 Russian by Jonatas Grosman
14
  results:
15
+ - task:
16
+ name: Automatic Speech Recognition
17
  type: automatic-speech-recognition
18
  dataset:
19
  name: Common Voice 8
20
  type: mozilla-foundation/common_voice_8_0
21
  args: ru
22
  metrics:
23
+ - name: Test WER
24
+ type: wer
25
+ value: 9.82
26
+ - name: Test CER
27
+ type: cer
28
+ value: 2.3
29
+ - name: Test WER (+LM)
30
+ type: wer
31
+ value: 7.08
32
+ - name: Test CER (+LM)
33
+ type: cer
34
+ value: 1.87
35
+ - task:
36
  name: Automatic Speech Recognition
37
  type: automatic-speech-recognition
38
  dataset:
40
  type: speech-recognition-community-v2/dev_data
41
  args: ru
42
  metrics:
43
+ - name: Dev WER
44
+ type: wer
45
+ value: 23.96
46
+ - name: Dev CER
47
+ type: cer
48
+ value: 8.88
49
+ - name: Dev WER (+LM)
50
+ type: wer
51
+ value: 15.88
52
+ - name: Dev CER (+LM)
53
+ type: cer
54
+ value: 7.42
55
+ - task:
56
+ name: Automatic Speech Recognition
57
+ type: automatic-speech-recognition
58
+ dataset:
59
+ name: Robust Speech Event - Test Data
60
+ type: speech-recognition-community-v2/eval_data
61
+ args: ru
62
+ metrics:
63
+ - name: Test WER
64
+ type: wer
65
+ value: 14.23
66
  ---
67
 
68
  # XLS-R-1B-RUSSIAN