Hugging Face's logo

ylacombe
/
wav2vec2-common_voice-tr-demo

Automatic Speech Recognition
Transformers
TensorBoard
Safetensors
Turkish
wav2vec2-bert
mozilla-foundation/common_voice_16_0
Generated from Trainer
Eval Results
Inference Endpoints
Model card Files Metrics Community
ylacombe HF staff commited on
Commit
221ab03
•
1 Parent(s): 38992cb

Model save

Browse files
Files changed (4) hide show
  1. README.md +59 -63
  2. emissions.csv +1 -0
  3. runs/Jan04_20-37-14_vorace/events.out.tfevents.1704401907.vorace.1537045.0 +2 -2
  4. runs/Jan05_08-39-35_vorace/events.out.tfevents.1704445083.vorace.1722965.0 +2 -2
README.md CHANGED
@@ -1,10 +1,6 @@
1
  ---
2
- language:
3
- - tr
4
  base_model: ylacombe/w2v-bert-2.0
5
  tags:
6
- - automatic-speech-recognition
7
- - mozilla-foundation/common_voice_16_0
8
  - generated_from_trainer
9
  datasets:
10
  - common_voice_16_0
@@ -17,11 +13,11 @@ model-index:
17
  name: Automatic Speech Recognition
18
  type: automatic-speech-recognition
19
  dataset:
20
- name: MOZILLA-FOUNDATION/COMMON_VOICE_16_0 - TR
21
  type: common_voice_16_0
22
  config: tr
23
  split: test
24
- args: 'Config: tr, Training split: train+validation, Eval split: test'
25
  metrics:
26
  - name: Wer
27
  type: wer
@@ -33,9 +29,9 @@ should probably proofread and complete it, then remove this comment. -->
33
 
34
  # wav2vec2-common_voice-tr-demo
35
 
36
- This model is a fine-tuned version of [ylacombe/w2v-bert-2.0](https://huggingface.co/ylacombe/w2v-bert-2.0) on the MOZILLA-FOUNDATION/COMMON_VOICE_16_0 - TR dataset.
37
  It achieves the following results on the evaluation set:
38
- - Loss: 3.0582
39
  - Wer: 1.0
40
 
41
  ## Model description
@@ -55,7 +51,7 @@ More information needed
55
  ### Training hyperparameters
56
 
57
  The following hyperparameters were used during training:
58
- - learning_rate: 0.00072018208512399
59
  - train_batch_size: 20
60
  - eval_batch_size: 32
61
  - seed: 42
@@ -71,60 +67,60 @@ The following hyperparameters were used during training:
71
 
72
  | Training Loss | Epoch | Step | Validation Loss | Wer |
73
  |:-------------:|:-----:|:-----:|:---------------:|:------:|
74
- | No log | 0.27 | 300 | 7.2663 | 1.0 |
75
- | 10.5256 | 0.55 | 600 | 3.0893 | 1.0 |
76
- | 10.5256 | 0.82 | 900 | 3.0612 | 1.0 |
77
- | 2.9795 | 1.1 | 1200 | 2.9937 | 1.0 |
78
- | 2.9564 | 1.37 | 1500 | 3.2424 | 1.0 |
79
- | 2.9564 | 1.64 | 1800 | 3.2866 | 1.0 |
80
- | 3.1552 | 1.92 | 2100 | 3.6339 | 1.0 |
81
- | 3.1552 | 2.19 | 2400 | 3.1185 | 1.0 |
82
- | 3.2079 | 2.47 | 2700 | 3.1832 | 1.0 |
83
- | 3.1275 | 2.74 | 3000 | 3.3952 | 1.0 |
84
- | 3.1275 | 3.01 | 3300 | 3.2982 | 1.0 |
85
- | 3.0987 | 3.29 | 3600 | 3.1036 | 1.0 |
86
- | 3.0987 | 3.56 | 3900 | 3.1223 | 1.0 |
87
- | 2.9301 | 3.84 | 4200 | 3.1145 | 1.0 |
88
- | 2.9197 | 4.11 | 4500 | 3.0324 | 1.0 |
89
- | 2.9197 | 4.38 | 4800 | 2.9994 | 1.9599 |
90
- | 2.9023 | 4.66 | 5100 | 2.9917 | 1.8240 |
91
- | 2.9023 | 4.93 | 5400 | 2.9946 | 1.9589 |
92
- | 2.9007 | 5.21 | 5700 | 3.1955 | 1.0 |
93
- | 3.1887 | 5.48 | 6000 | 3.1902 | 1.0 |
94
- | 3.1887 | 5.75 | 6300 | 3.1672 | 1.0 |
95
- | 3.135 | 6.03 | 6600 | 3.2076 | 1.0 |
96
- | 3.135 | 6.3 | 6900 | 3.2120 | 1.0 |
97
- | 3.1482 | 6.58 | 7200 | 3.1832 | 1.0 |
98
- | 3.1546 | 6.85 | 7500 | 3.1799 | 1.0 |
99
- | 3.1546 | 7.12 | 7800 | 3.2452 | 1.0 |
100
- | 3.1567 | 7.4 | 8100 | 3.2319 | 1.0 |
101
- | 3.1567 | 7.67 | 8400 | 3.2228 | 1.0 |
102
- | 3.1719 | 7.95 | 8700 | 3.2055 | 1.0 |
103
- | 3.168 | 8.22 | 9000 | 3.2553 | 1.0 |
104
- | 3.168 | 8.49 | 9300 | 3.1975 | 1.0 |
105
- | 3.1643 | 8.77 | 9600 | 3.2446 | 1.0 |
106
- | 3.1643 | 9.04 | 9900 | 3.2781 | 1.0 |
107
- | 3.169 | 9.32 | 10200 | 3.2597 | 1.0 |
108
- | 3.1789 | 9.59 | 10500 | 3.2586 | 1.0 |
109
- | 3.1789 | 9.86 | 10800 | 3.2690 | 1.0 |
110
- | 3.1701 | 10.14 | 11100 | 3.2737 | 1.0 |
111
- | 3.1701 | 10.41 | 11400 | 3.2738 | 1.0 |
112
- | 3.1698 | 10.68 | 11700 | 3.2595 | 1.0 |
113
- | 3.1595 | 10.96 | 12000 | 3.2467 | 1.0 |
114
- | 3.1595 | 11.23 | 12300 | 3.2524 | 1.0 |
115
- | 3.15 | 11.51 | 12600 | 3.2327 | 1.0 |
116
- | 3.15 | 11.78 | 12900 | 3.2196 | 1.0 |
117
- | 3.1444 | 12.05 | 13200 | 3.1943 | 1.0 |
118
- | 3.132 | 12.33 | 13500 | 3.1911 | 1.0 |
119
- | 3.132 | 12.6 | 13800 | 3.2075 | 1.0 |
120
- | 3.1153 | 12.88 | 14100 | 3.1938 | 1.0 |
121
- | 3.1153 | 13.15 | 14400 | 3.1639 | 1.0 |
122
- | 3.1039 | 13.42 | 14700 | 3.1515 | 1.0 |
123
- | 3.0839 | 13.7 | 15000 | 3.1535 | 1.0 |
124
- | 3.0839 | 13.97 | 15300 | 3.1307 | 1.0 |
125
- | 3.0632 | 14.25 | 15600 | 3.1138 | 1.0 |
126
- | 3.0632 | 14.52 | 15900 | 3.1289 | 1.0 |
127
- | 3.0518 | 14.79 | 16200 | 3.0815 | 1.0 |
128
 
129
 
130
  ### Framework versions
 
1
  ---
 
 
2
  base_model: ylacombe/w2v-bert-2.0
3
  tags:
 
 
4
  - generated_from_trainer
5
  datasets:
6
  - common_voice_16_0
 
13
  name: Automatic Speech Recognition
14
  type: automatic-speech-recognition
15
  dataset:
16
+ name: common_voice_16_0
17
  type: common_voice_16_0
18
  config: tr
19
  split: test
20
+ args: tr
21
  metrics:
22
  - name: Wer
23
  type: wer
 
29
 
30
  # wav2vec2-common_voice-tr-demo
31
 
32
+ This model is a fine-tuned version of [ylacombe/w2v-bert-2.0](https://huggingface.co/ylacombe/w2v-bert-2.0) on the common_voice_16_0 dataset.
33
  It achieves the following results on the evaluation set:
34
+ - Loss: nan
35
  - Wer: 1.0
36
 
37
  ## Model description
 
51
  ### Training hyperparameters
52
 
53
  The following hyperparameters were used during training:
54
+ - learning_rate: 0.007448827845832091
55
  - train_batch_size: 20
56
  - eval_batch_size: 32
57
  - seed: 42
 
67
 
68
  | Training Loss | Epoch | Step | Validation Loss | Wer |
69
  |:-------------:|:-----:|:-----:|:---------------:|:------:|
70
+ | No log | 0.27 | 300 | 3.2930 | 1.0 |
71
+ | 5.6462 | 0.55 | 600 | 3.4159 | 1.0 |
72
+ | 5.6462 | 0.82 | 900 | 3.4422 | 1.0 |
73
+ | 3.3522 | 1.1 | 1200 | 3.3719 | 1.0 |
74
+ | 3.2605 | 1.37 | 1500 | 3.4026 | 1.0 |
75
+ | 3.2605 | 1.64 | 1800 | 3.4448 | 1.0 |
76
+ | 3.2766 | 1.92 | 2100 | 3.4736 | 0.9999 |
77
+ | 3.2766 | 2.19 | 2400 | 3.9828 | 1.0 |
78
+ | 3.2853 | 2.47 | 2700 | 3.5532 | 1.0 |
79
+ | 3.3389 | 2.74 | 3000 | 3.7819 | 1.0 |
80
+ | 3.3389 | 3.01 | 3300 | 3.2250 | 1.0 |
81
+ | 3.2186 | 3.29 | 3600 | 3.2373 | 1.0 |
82
+ | 3.2186 | 3.56 | 3900 | 3.2162 | 1.0 |
83
+ | 3.1916 | 3.84 | 4200 | 3.2368 | 1.0 |
84
+ | 3.2188 | 4.11 | 4500 | 3.2377 | 1.0 |
85
+ | 3.2188 | 4.38 | 4800 | 3.4207 | 1.0 |
86
+ | 5.3067 | 4.66 | 5100 | nan | 1.0 |
87
+ | 5.3067 | 4.93 | 5400 | nan | 1.0 |
88
+ | 0.0 | 5.21 | 5700 | nan | 1.0 |
89
+ | 0.0 | 5.48 | 6000 | nan | 1.0 |
90
+ | 0.0 | 5.75 | 6300 | nan | 1.0 |
91
+ | 0.0 | 6.03 | 6600 | nan | 1.0 |
92
+ | 0.0 | 6.3 | 6900 | nan | 1.0 |
93
+ | 0.0 | 6.58 | 7200 | nan | 1.0 |
94
+ | 0.0 | 6.85 | 7500 | nan | 1.0 |
95
+ | 0.0 | 7.12 | 7800 | nan | 1.0 |
96
+ | 0.0 | 7.4 | 8100 | nan | 1.0 |
97
+ | 0.0 | 7.67 | 8400 | nan | 1.0 |
98
+ | 0.0 | 7.95 | 8700 | nan | 1.0 |
99
+ | 0.0 | 8.22 | 9000 | nan | 1.0 |
100
+ | 0.0 | 8.49 | 9300 | nan | 1.0 |
101
+ | 0.0 | 8.77 | 9600 | nan | 1.0 |
102
+ | 0.0 | 9.04 | 9900 | nan | 1.0 |
103
+ | 0.0 | 9.32 | 10200 | nan | 1.0 |
104
+ | 0.0 | 9.59 | 10500 | nan | 1.0 |
105
+ | 0.0 | 9.86 | 10800 | nan | 1.0 |
106
+ | 0.0 | 10.14 | 11100 | nan | 1.0 |
107
+ | 0.0 | 10.41 | 11400 | nan | 1.0 |
108
+ | 0.0 | 10.68 | 11700 | nan | 1.0 |
109
+ | 0.0 | 10.96 | 12000 | nan | 1.0 |
110
+ | 0.0 | 11.23 | 12300 | nan | 1.0 |
111
+ | 0.0 | 11.51 | 12600 | nan | 1.0 |
112
+ | 0.0 | 11.78 | 12900 | nan | 1.0 |
113
+ | 0.0 | 12.05 | 13200 | nan | 1.0 |
114
+ | 0.0 | 12.33 | 13500 | nan | 1.0 |
115
+ | 0.0 | 12.6 | 13800 | nan | 1.0 |
116
+ | 0.0 | 12.88 | 14100 | nan | 1.0 |
117
+ | 0.0 | 13.15 | 14400 | nan | 1.0 |
118
+ | 0.0 | 13.42 | 14700 | nan | 1.0 |
119
+ | 0.0 | 13.7 | 15000 | nan | 1.0 |
120
+ | 0.0 | 13.97 | 15300 | nan | 1.0 |
121
+ | 0.0 | 14.25 | 15600 | nan | 1.0 |
122
+ | 0.0 | 14.52 | 15900 | nan | 1.0 |
123
+ | 0.0 | 14.79 | 16200 | nan | 1.0 |
124
 
125
 
126
  ### Framework versions
emissions.csv CHANGED
@@ -1,2 +1,3 @@
1
  timestamp,experiment_id,project_name,duration,emissions,energy_consumed,country_name,country_iso_code,region,on_cloud,cloud_provider,cloud_region
2
  2024-01-05T08:32:01,fd2c0c3c-2242-464b-88fe-382a042b9afd,codecarbon,44964.32544207573,2.608847954164164,6.14448116064,France,FRA,île-de-france,N,,
 
 
1
  timestamp,experiment_id,project_name,duration,emissions,energy_consumed,country_name,country_iso_code,region,on_cloud,cloud_provider,cloud_region
2
  2024-01-05T08:32:01,fd2c0c3c-2242-464b-88fe-382a042b9afd,codecarbon,44964.32544207573,2.608847954164164,6.14448116064,France,FRA,île-de-france,N,,
3
+ 2024-01-05T09:39:18,5021c317-a2d1-462c-9508-e6b1ff4e2091,codecarbon,45648.984209775925,2.6037451682098878,6.1324628396363625,France,FRA,île-de-france,N,,
runs/Jan04_20-37-14_vorace/events.out.tfevents.1704401907.vorace.1537045.0 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:e0378b87a0daed020ebe2efac3bbfdc4c58ed6b44cc22de7b0b89491563f52e5
3
- size 27845
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:20102dc891b75498f662e4396918dfc6742689f64ec0b45b338c1f00944dd183
3
+ size 28205
runs/Jan05_08-39-35_vorace/events.out.tfevents.1704445083.vorace.1722965.0 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:8512187ad90368d0a342a4df67b83eb2d77ff8579a7da68318edcd9e4ad9db4e
3
- size 6443
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d83369742a1f74133bc19aca815f10b018697dbe547b6f55b85b1910cbc89544
3
+ size 6761