Automatic Speech Recognition
Transformers
PyTorch
TensorBoard
French
whisper
hf-asr-leaderboard
whisper-event
Eval Results
Inference Endpoints
librarian-bot commited on
Commit
d5950ce
1 Parent(s): f3b39d6

Librarian Bot: Add base_model information to model

Browse files

This pull request aims to enrich the metadata of your model by adding [`openai/whisper-large-v2`](https://huggingface.co/openai/whisper-large-v2) as a `base_model` field, situated in the `YAML` block of your model's `README.md`.

How did we find this information? We performed a regular expression match on your `README.md` file to determine the connection.

**Why add this?** Enhancing your model's metadata in this way:
- **Boosts Discoverability** - It becomes straightforward to trace the relationships between various models on the Hugging Face Hub.
- **Highlights Impact** - It showcases the contributions and influences different models have within the community.

For a hands-on example of how such metadata can play a pivotal role in mapping model connections, take a look at [librarian-bots/base_model_explorer](https://huggingface.co/spaces/librarian-bots/base_model_explorer).

This PR comes courtesy of [Librarian Bot](https://huggingface.co/librarian-bot). If you have any feedback, queries, or need assistance, please don't hesitate to reach out to [@davanstrien](https://huggingface.co/davanstrien). Your input is invaluable to us!

Files changed (1) hide show
  1. README.md +29 -29
README.md CHANGED
@@ -1,8 +1,7 @@
1
  ---
2
- license: apache-2.0
3
  language: fr
 
4
  library_name: transformers
5
- thumbnail: null
6
  tags:
7
  - automatic-speech-recognition
8
  - hf-asr-leaderboard
@@ -15,12 +14,13 @@ datasets:
15
  - gigant/african_accented_french
16
  metrics:
17
  - wer
 
18
  model-index:
19
  - name: Fine-tuned whisper-large-v2 model for ASR in French
20
  results:
21
  - task:
22
- name: Automatic Speech Recognition
23
  type: automatic-speech-recognition
 
24
  dataset:
25
  name: Common Voice 11.0
26
  type: mozilla-foundation/common_voice_11_0
@@ -28,15 +28,15 @@ model-index:
28
  split: test
29
  args: fr
30
  metrics:
31
- - name: WER (Greedy)
32
- type: wer
33
  value: 8.15
34
- - name: WER (Beam 5)
35
- type: wer
36
  value: 7.83
 
37
  - task:
38
- name: Automatic Speech Recognition
39
  type: automatic-speech-recognition
 
40
  dataset:
41
  name: Multilingual LibriSpeech (MLS)
42
  type: facebook/multilingual_librispeech
@@ -44,15 +44,15 @@ model-index:
44
  split: test
45
  args: french
46
  metrics:
47
- - name: WER (Greedy)
48
- type: wer
49
- value: 4.20
50
- - name: WER (Beam 5)
51
- type: wer
52
  value: 4.03
 
53
  - task:
54
- name: Automatic Speech Recognition
55
  type: automatic-speech-recognition
 
56
  dataset:
57
  name: VoxPopuli
58
  type: facebook/voxpopuli
@@ -60,15 +60,15 @@ model-index:
60
  split: test
61
  args: fr
62
  metrics:
63
- - name: WER (Greedy)
64
- type: wer
65
- value: 9.10
66
- - name: WER (Beam 5)
67
- type: wer
68
  value: 8.66
 
69
  - task:
70
- name: Automatic Speech Recognition
71
  type: automatic-speech-recognition
 
72
  dataset:
73
  name: Fleurs
74
  type: google/fleurs
@@ -76,15 +76,15 @@ model-index:
76
  split: test
77
  args: fr_fr
78
  metrics:
79
- - name: WER (Greedy)
80
- type: wer
81
  value: 5.22
82
- - name: WER (Beam 5)
83
- type: wer
84
  value: 4.98
 
85
  - task:
86
- name: Automatic Speech Recognition
87
  type: automatic-speech-recognition
 
88
  dataset:
89
  name: African Accented French
90
  type: gigant/african_accented_french
@@ -92,12 +92,12 @@ model-index:
92
  split: test
93
  args: fr
94
  metrics:
95
- - name: WER (Greedy)
96
- type: wer
97
  value: 4.58
98
- - name: WER (Beam 5)
99
- type: wer
100
  value: 4.31
 
101
  ---
102
 
103
  <style>
 
1
  ---
 
2
  language: fr
3
+ license: apache-2.0
4
  library_name: transformers
 
5
  tags:
6
  - automatic-speech-recognition
7
  - hf-asr-leaderboard
 
14
  - gigant/african_accented_french
15
  metrics:
16
  - wer
17
+ base_model: openai/whisper-large-v2
18
  model-index:
19
  - name: Fine-tuned whisper-large-v2 model for ASR in French
20
  results:
21
  - task:
 
22
  type: automatic-speech-recognition
23
+ name: Automatic Speech Recognition
24
  dataset:
25
  name: Common Voice 11.0
26
  type: mozilla-foundation/common_voice_11_0
 
28
  split: test
29
  args: fr
30
  metrics:
31
+ - type: wer
 
32
  value: 8.15
33
+ name: WER (Greedy)
34
+ - type: wer
35
  value: 7.83
36
+ name: WER (Beam 5)
37
  - task:
 
38
  type: automatic-speech-recognition
39
+ name: Automatic Speech Recognition
40
  dataset:
41
  name: Multilingual LibriSpeech (MLS)
42
  type: facebook/multilingual_librispeech
 
44
  split: test
45
  args: french
46
  metrics:
47
+ - type: wer
48
+ value: 4.2
49
+ name: WER (Greedy)
50
+ - type: wer
 
51
  value: 4.03
52
+ name: WER (Beam 5)
53
  - task:
 
54
  type: automatic-speech-recognition
55
+ name: Automatic Speech Recognition
56
  dataset:
57
  name: VoxPopuli
58
  type: facebook/voxpopuli
 
60
  split: test
61
  args: fr
62
  metrics:
63
+ - type: wer
64
+ value: 9.1
65
+ name: WER (Greedy)
66
+ - type: wer
 
67
  value: 8.66
68
+ name: WER (Beam 5)
69
  - task:
 
70
  type: automatic-speech-recognition
71
+ name: Automatic Speech Recognition
72
  dataset:
73
  name: Fleurs
74
  type: google/fleurs
 
76
  split: test
77
  args: fr_fr
78
  metrics:
79
+ - type: wer
 
80
  value: 5.22
81
+ name: WER (Greedy)
82
+ - type: wer
83
  value: 4.98
84
+ name: WER (Beam 5)
85
  - task:
 
86
  type: automatic-speech-recognition
87
+ name: Automatic Speech Recognition
88
  dataset:
89
  name: African Accented French
90
  type: gigant/african_accented_french
 
92
  split: test
93
  args: fr
94
  metrics:
95
+ - type: wer
 
96
  value: 4.58
97
+ name: WER (Greedy)
98
+ - type: wer
99
  value: 4.31
100
+ name: WER (Beam 5)
101
  ---
102
 
103
  <style>