Hugging Face's logo

Kyro-2023
/
wav2vec2-large-mms-1b-turkish-colab

Automatic Speech Recognition
Transformers
PyTorch
wav2vec2
Generated from Trainer
Eval Results
Inference Endpoints
Model card Files Community
Kyro-2023 commited on
Commit
a4ad48d
•
1 Parent(s): d8c2584

End of training

Browse files
Files changed (3) hide show
  1. README.md +12 -5
  2. adapter.tur.safetensors +1 -1
  3. pytorch_model.bin +1 -1
README.md CHANGED
@@ -22,7 +22,7 @@ model-index:
22
  metrics:
23
  - name: Wer
24
  type: wer
25
- value: 0.24471453375548974
26
  ---
27
 
28
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -32,8 +32,8 @@ should probably proofread and complete it, then remove this comment. -->
32
 
33
  This model is a fine-tuned version of [facebook/mms-1b-all](https://huggingface.co/facebook/mms-1b-all) on the common_voice_6_1 dataset.
34
  It achieves the following results on the evaluation set:
35
- - Loss: 0.1785
36
- - Wer: 0.2447
37
 
38
  ## Model description
39
 
@@ -59,13 +59,20 @@ The following hyperparameters were used during training:
59
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
60
  - lr_scheduler_type: linear
61
  - lr_scheduler_warmup_steps: 100
62
- - num_epochs: 1
63
 
64
  ### Training results
65
 
66
  | Training Loss | Epoch | Step | Validation Loss | Wer |
67
  |:-------------:|:-----:|:----:|:---------------:|:------:|
68
- | 4.0431 | 0.92 | 100 | 0.1785 | 0.2447 |
 
 
 
 
 
 
 
69
 
70
 
71
  ### Framework versions
 
22
  metrics:
23
  - name: Wer
24
  type: wer
25
+ value: 0.21101011132672862
26
  ---
27
 
28
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 
32
 
33
  This model is a fine-tuned version of [facebook/mms-1b-all](https://huggingface.co/facebook/mms-1b-all) on the common_voice_6_1 dataset.
34
  It achieves the following results on the evaluation set:
35
+ - Loss: 0.1472
36
+ - Wer: 0.2110
37
 
38
  ## Model description
39
 
 
59
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
60
  - lr_scheduler_type: linear
61
  - lr_scheduler_warmup_steps: 100
62
+ - num_epochs: 8
63
 
64
  ### Training results
65
 
66
  | Training Loss | Epoch | Step | Validation Loss | Wer |
67
  |:-------------:|:-----:|:----:|:---------------:|:------:|
68
+ | 4.6395 | 0.92 | 100 | 0.1800 | 0.2494 |
69
+ | 0.2845 | 1.83 | 200 | 0.1673 | 0.2354 |
70
+ | 0.2692 | 2.75 | 300 | 0.1573 | 0.2227 |
71
+ | 0.245 | 3.67 | 400 | 0.1568 | 0.2147 |
72
+ | 0.2385 | 4.59 | 500 | 0.1533 | 0.2164 |
73
+ | 0.2416 | 5.5 | 600 | 0.1502 | 0.2139 |
74
+ | 0.2182 | 6.42 | 700 | 0.1507 | 0.2124 |
75
+ | 0.2276 | 7.34 | 800 | 0.1472 | 0.2110 |
76
 
77
 
78
  ### Framework versions
adapter.tur.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:580137006845e3950f8a22b11dc40b6aea0a28226de9333c142a43e452859181
3
  size 8839532
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:33f1cf6cff5ca5803ba83da2d4edb8dd7e94c5d2f591b1c1c4d85e6c20ed46ad
3
  size 8839532
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:6303f862bef91823b3e68be274663934284915232aa5acd08d0f387e1b773558
3
  size 3859175565
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6ecd68333cf4643b0d374c12c3c6a14348712bf0391bcfaf5c5ee2f58987b551
3
  size 3859175565