Hugging Face's logo

kotoba-tech
/
kotoba-whisper-v2.0

Automatic Speech Recognition
Transformers
Safetensors
Japanese
whisper
audio
hf-asr-leaderboard
Eval Results
Inference Endpoints
Model card Files Community
asahi417 commited on
Commit
1f82629
•
1 Parent(s): f1d4517

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +43 -38
README.md CHANGED
@@ -18,45 +18,50 @@ widget:
18
  pipeline_tag: automatic-speech-recognition
19
  metrics:
20
  - wer
 
21
  model-index:
22
- - name: kotoba-tech/kotoba-whisper-v2.0
23
- results:
24
- - task:
25
- type: automatic-speech-recognition
26
- dataset:
27
- name: CommonVoice_8.0 (Japanese)
28
- type: japanese-asr/ja_asr.common_voice_8_0
29
- metrics:
30
- - name: WER
31
- type: WER
32
- value: 58.9
33
- - name: CER
34
- type: CER
35
- value: 9.20
36
- - task:
37
- type: automatic-speech-recognition
38
- dataset:
39
- name: ReazonSpeech (Test)
40
- type: japanese-asr/ja_asr.reazonspeech_test
41
- metrics:
42
- - name: WER
43
- type: WER
44
- value: 55.6
45
- - name: CER
46
- type: CER
47
- value: 11.63
48
- - task:
49
- type: automatic-speech-recognition
50
- dataset:
51
- name: JSUT Basic5000
52
- type: japanese-asr/ja_asr.jsut_basic5000
53
- metrics:
54
- - name: WER
55
- type: WER
56
- value: 63.8
57
- - name: CER
58
- type: CER
59
- value: 8.40
 
 
 
 
60
  ---
61
 
62
  # Kotoba-Whisper (v2.0)
 
18
  pipeline_tag: automatic-speech-recognition
19
  metrics:
20
  - wer
21
+ - cer
22
  model-index:
23
+ - name: kotoba-tech/kotoba-whisper-v2.0
24
+ results:
25
+ - task:
26
+ type: automatic-speech-recognition
27
+ dataset:
28
+ name: CommonVoice_8.0 (Japanese)
29
+ type: japanese-asr/ja_asr.common_voice_8_0
30
+ metrics:
31
+ - name: WER
32
+ type: WER
33
+ value: 58.9
34
+ - name: CER
35
+ type: CER
36
+ value: 9.2
37
+ - task:
38
+ type: automatic-speech-recognition
39
+ dataset:
40
+ name: ReazonSpeech (Test)
41
+ type: japanese-asr/ja_asr.reazonspeech_test
42
+ metrics:
43
+ - name: WER
44
+ type: WER
45
+ value: 55.6
46
+ - name: CER
47
+ type: CER
48
+ value: 11.63
49
+ - task:
50
+ type: automatic-speech-recognition
51
+ dataset:
52
+ name: JSUT Basic5000
53
+ type: japanese-asr/ja_asr.jsut_basic5000
54
+ metrics:
55
+ - name: WER
56
+ type: WER
57
+ value: 63.8
58
+ - name: CER
59
+ type: CER
60
+ value: 8.4
61
+ datasets:
62
+ - japanese-asr/whisper_transcriptions.reazonspeech.all
63
+ - japanese-asr/whisper_transcriptions.reazonspeech.all.wer_10.0
64
+ - japanese-asr/whisper_transcriptions.reazonspeech.all.wer_10.0.vectorized
65
  ---
66
 
67
  # Kotoba-Whisper (v2.0)