Hugging Face's logo

ndeclarke
/
wav2vec2-mms-1b-CV17.0-training_set_variations

Automatic Speech Recognition
Transformers
TensorBoard
Safetensors
wav2vec2
Generated from Trainer
Eval Results
Inference Endpoints
Model card Files Metrics Community
ndeclarke commited on
Commit
f9f86aa
•
1 Parent(s): 6492a06

End of training

Browse files
Files changed (3) hide show
  1. README.md +30 -37
  2. adapter.tam-32.safetensors +1 -1
  3. model.safetensors +1 -1
README.md CHANGED
@@ -1,20 +1,20 @@
1
  ---
 
 
2
  base_model: facebook/mms-1b-all
 
 
3
  datasets:
4
  - common_voice_17_0
5
- library_name: transformers
6
- license: cc-by-nc-4.0
7
  metrics:
8
  - wer
9
  - bleu
10
- tags:
11
- - generated_from_trainer
12
  model-index:
13
  - name: wav2vec2-mms-1b-CV17.0-training_set_variations
14
  results:
15
  - task:
16
- type: automatic-speech-recognition
17
  name: Automatic Speech Recognition
 
18
  dataset:
19
  name: common_voice_17_0
20
  type: common_voice_17_0
@@ -22,12 +22,12 @@ model-index:
22
  split: validation
23
  args: ta
24
  metrics:
25
- - type: wer
26
- value: 0.35028609973948643
27
- name: Wer
28
- - type: bleu
29
- value: 0.4335646189418293
30
- name: Bleu
31
  ---
32
 
33
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -37,10 +37,10 @@ should probably proofread and complete it, then remove this comment. -->
37
 
38
  This model is a fine-tuned version of [facebook/mms-1b-all](https://huggingface.co/facebook/mms-1b-all) on the common_voice_17_0 dataset.
39
  It achieves the following results on the evaluation set:
40
- - Loss: 0.1970
41
- - Wer: 0.3503
42
- - Cer: 0.0557
43
- - Bleu: 0.4336
44
 
45
  ## Model description
46
 
@@ -73,28 +73,21 @@ The following hyperparameters were used during training:
73
 
74
  ### Training results
75
 
76
- | Training Loss | Epoch | Step | Validation Loss | Wer | Cer | Bleu |
77
- |:-------------:|:------:|:----:|:---------------:|:------:|:------:|:------:|
78
- | 7.329 | 0.0977 | 100 | 0.3217 | 0.4338 | 0.0716 | 0.3272 |
79
- | 0.2154 | 0.1953 | 200 | 0.2314 | 0.3794 | 0.0634 | 0.3999 |
80
- | 0.189 | 0.2930 | 300 | 0.2188 | 0.3656 | 0.0592 | 0.4166 |
81
- | 0.1975 | 0.3906 | 400 | 0.2212 | 0.3763 | 0.0608 | 0.4032 |
82
- | 0.1813 | 0.4883 | 500 | 0.2117 | 0.3634 | 0.0585 | 0.4171 |
83
- | 0.1791 | 0.5859 | 600 | 0.2074 | 0.3590 | 0.0578 | 0.4220 |
84
- | 0.187 | 0.6836 | 700 | 0.2087 | 0.3607 | 0.0582 | 0.4188 |
85
- | 0.1789 | 0.7812 | 800 | 0.2064 | 0.3542 | 0.0568 | 0.4327 |
86
- | 0.1704 | 0.8789 | 900 | 0.2076 | 0.3661 | 0.0587 | 0.4095 |
87
- | 0.1813 | 0.9766 | 1000 | 0.2044 | 0.3589 | 0.0574 | 0.4200 |
88
- | 0.1633 | 1.0742 | 1100 | 0.2029 | 0.3582 | 0.0575 | 0.4220 |
89
- | 0.1699 | 1.1719 | 1200 | 0.2034 | 0.3537 | 0.0566 | 0.4335 |
90
- | 0.1822 | 1.2695 | 1300 | 0.2037 | 0.3589 | 0.0578 | 0.4227 |
91
- | 0.1654 | 1.3672 | 1400 | 0.2028 | 0.3549 | 0.0568 | 0.4288 |
92
- | 0.1696 | 1.4648 | 1500 | 0.2011 | 0.3579 | 0.0567 | 0.4199 |
93
- | 0.1622 | 1.5625 | 1600 | 0.1999 | 0.3568 | 0.0570 | 0.4228 |
94
- | 0.1742 | 1.6602 | 1700 | 0.1983 | 0.3490 | 0.0559 | 0.4365 |
95
- | 0.1581 | 1.7578 | 1800 | 0.1973 | 0.3511 | 0.0558 | 0.4329 |
96
- | 0.1616 | 1.8555 | 1900 | 0.1970 | 0.3482 | 0.0556 | 0.4381 |
97
- | 0.1607 | 1.9531 | 2000 | 0.1970 | 0.3503 | 0.0557 | 0.4336 |
98
 
99
 
100
  ### Framework versions
 
1
  ---
2
+ library_name: transformers
3
+ license: cc-by-nc-4.0
4
  base_model: facebook/mms-1b-all
5
+ tags:
6
+ - generated_from_trainer
7
  datasets:
8
  - common_voice_17_0
 
 
9
  metrics:
10
  - wer
11
  - bleu
 
 
12
  model-index:
13
  - name: wav2vec2-mms-1b-CV17.0-training_set_variations
14
  results:
15
  - task:
 
16
  name: Automatic Speech Recognition
17
+ type: automatic-speech-recognition
18
  dataset:
19
  name: common_voice_17_0
20
  type: common_voice_17_0
 
22
  split: validation
23
  args: ta
24
  metrics:
25
+ - name: Wer
26
+ type: wer
27
+ value: 0.4958060228262364
28
+ - name: Bleu
29
+ type: bleu
30
+ value: 0.2629057639184852
31
  ---
32
 
33
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 
37
 
38
  This model is a fine-tuned version of [facebook/mms-1b-all](https://huggingface.co/facebook/mms-1b-all) on the common_voice_17_0 dataset.
39
  It achieves the following results on the evaluation set:
40
+ - Loss: 0.5099
41
+ - Wer: 0.4958
42
+ - Cer: 0.0885
43
+ - Bleu: 0.2629
44
 
45
  ## Model description
46
 
 
73
 
74
  ### Training results
75
 
76
+ | Training Loss | Epoch | Step | Validation Loss | Wer | Cer | Bleu |
77
+ |:-------------:|:-----:|:----:|:---------------:|:------:|:------:|:------:|
78
+ | 11.4061 | 50.0 | 50 | 4.6415 | 1.0007 | 0.9640 | 0.0 |
79
+ | 1.789 | 100.0 | 100 | 0.3026 | 0.4457 | 0.0734 | 0.3064 |
80
+ | 0.08 | 150.0 | 150 | 0.3223 | 0.4304 | 0.0711 | 0.3275 |
81
+ | 0.0473 | 200.0 | 200 | 0.3547 | 0.4426 | 0.0742 | 0.3156 |
82
+ | 0.0364 | 250.0 | 250 | 0.3786 | 0.4556 | 0.0761 | 0.2972 |
83
+ | 0.0298 | 300.0 | 300 | 0.4070 | 0.4629 | 0.0800 | 0.2875 |
84
+ | 0.0279 | 350.0 | 350 | 0.4190 | 0.4688 | 0.0799 | 0.2864 |
85
+ | 0.0253 | 400.0 | 400 | 0.4353 | 0.4755 | 0.0818 | 0.2757 |
86
+ | 0.0198 | 450.0 | 450 | 0.4808 | 0.5066 | 0.0887 | 0.2432 |
87
+ | 0.0216 | 500.0 | 500 | 0.4699 | 0.4780 | 0.0815 | 0.2777 |
88
+ | 0.0194 | 550.0 | 550 | 0.4745 | 0.4895 | 0.0877 | 0.2643 |
89
+ | 0.0201 | 600.0 | 600 | 0.5035 | 0.4971 | 0.0881 | 0.2647 |
90
+ | 0.0153 | 650.0 | 650 | 0.5099 | 0.4958 | 0.0885 | 0.2629 |
 
 
 
 
 
 
 
91
 
92
 
93
  ### Framework versions
adapter.tam-32.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:4f8fe71b9a7e8a2606159ed838c80f25bd4a6762194f6030cf29d4d41fd79a25
3
  size 8926640
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1fb2b1bdfdf29c221dcb428919b9c1908a86f7ec6e120d96b938ac552b4bc0ba
3
  size 8926640
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:2e15947ff491a2ffd5857ec1256b03d6baa0317c255c1b42db33e65d5db55563
3
  size 3859019024
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ee9bb620c3c54a573ec782f35d6b1a5e34cfe5f750b5783967fb2f8b486a64b3
3
  size 3859019024