Hugging Face's logo

devasheeshG
/
whisper_medium_fp16_transformers

Automatic Speech Recognition
Transformers
PyTorch
whisper
audio
speech
wav2vec2
Eval Results
Model card Files
xet
Community
1
devasheeshG commited on
Commit
b656fa6
·
1 Parent(s): fbc6b69

.

Browse files
Files changed (1) hide show
  1. README.md +36 -36
README.md CHANGED
@@ -20,53 +20,53 @@ model-index:
20
  name: Common Voice (14.0) (Hindi) (test.tsv -> 2557 samples used)
21
  metrics:
22
  - type: wer
23
- value: 1.7
24
  name: Test WER
25
  description: Word Error Rate
26
  - type: mer
27
- value: 1.1
28
  name: Test MER
29
  description: Match Error Rate
30
  - type: wil
31
- value: 3,584
32
  name: Test WIL
33
  description: Word Information Lost
34
  - type: wip
35
- value: 112
36
  name: Test WIP
37
  description: Word Information Preserved
38
  - type: cer
39
- value: 1.7
40
  name: Test CER
41
  description: Character Error Rate
42
 
43
- - task:
44
- type: automatic-speech-recognition
45
- name: Automatic Speech Recognition
46
- dataset:
47
- type: common_voice
48
- name: Common Voice (14.0) (English) (test.tsv -> 2557 samples used)
49
- metrics:
50
- - type: wer
51
- value: -
52
- name: Test WER
53
- description: Word Error Rate
54
- - type: mer
55
- value: -
56
- name: Test MER
57
- description: Match Error Rate
58
- - type: wil
59
- value: -
60
- name: Test WIL
61
- description: Word Information Lost
62
- - type: wip
63
- value: -
64
- name: Test WIP
65
- description: Word Information Preserved
66
- - type: cer
67
- value: -
68
- name: Test CER
69
- description: Character Error Rate
70
 
71
  widget:
72
  - example_title: Librispeech sample 1
@@ -216,10 +216,10 @@ language:
216
  - **CER: Character Error Rate**
217
 
218
  ### Hindi (test.tsv -> 2557 samples used) [Common Voice 14.0](https://commonvoice.mozilla.org/en/datasets)
219
- | | WER | MER | WIL | WIP | CER |
220
- | ----------------- | -------------------- | ------- | --------- | ----------- | --- |
221
- | Original_Model | - | - | - | - | - |
222
- | This_Model | - | - | - | - | - |
223
 
224
  ### English
225
  | | WER | MER | WIL | WIP | CER |
 
20
  name: Common Voice (14.0) (Hindi) (test.tsv -> 2557 samples used)
21
  metrics:
22
  - type: wer
23
+ value: 54.97
24
  name: Test WER
25
  description: Word Error Rate
26
  - type: mer
27
+ value: 47.86
28
  name: Test MER
29
  description: Match Error Rate
30
  - type: wil
31
+ value: 66.83
32
  name: Test WIL
33
  description: Word Information Lost
34
  - type: wip
35
+ value: 33.16
36
  name: Test WIP
37
  description: Word Information Preserved
38
  - type: cer
39
+ value: 30.23
40
  name: Test CER
41
  description: Character Error Rate
42
 
43
+ # - task:
44
+ # type: automatic-speech-recognition
45
+ # name: Automatic Speech Recognition
46
+ # dataset:
47
+ # type: common_voice
48
+ # name: Common Voice (14.0) (English) (test.tsv -> 2557 samples used)
49
+ # metrics:
50
+ # - type: wer
51
+ # value: -
52
+ # name: Test WER
53
+ # description: Word Error Rate
54
+ # - type: mer
55
+ # value: -
56
+ # name: Test MER
57
+ # description: Match Error Rate
58
+ # - type: wil
59
+ # value: -
60
+ # name: Test WIL
61
+ # description: Word Information Lost
62
+ # - type: wip
63
+ # value: -
64
+ # name: Test WIP
65
+ # description: Word Information Preserved
66
+ # - type: cer
67
+ # value: -
68
+ # name: Test CER
69
+ # description: Character Error Rate
70
 
71
  widget:
72
  - example_title: Librispeech sample 1
 
216
  - **CER: Character Error Rate**
217
 
218
  ### Hindi (test.tsv -> 2557 samples used) [Common Voice 14.0](https://commonvoice.mozilla.org/en/datasets)
219
+ | | WER | MER | WIL | WIP | CER |
220
+ | ----------------- | -------------------- | ------- | --------- | ----------- | ---- |
221
+ | Original_Model | - | - | - | - | - |
222
+ | This_Model | 54.97 | 47.86 | 66.83 | 33.16 | 30.23|
223
 
224
  ### English
225
  | | WER | MER | WIL | WIP | CER |