Hugging Face's logo

nutella-toast
/
wav2vec2-large-xls-r-ssw

Automatic Speech Recognition
Transformers
TensorBoard
Safetensors
wav2vec2
Generated from Trainer
Eval Results
Inference Endpoints
Model card Files Metrics Community
nutella-toast commited on
Commit
b3a5e81
•
1 Parent(s): 8c4e9b0

End of training

Browse files
Files changed (3) hide show
  1. README.md +27 -18
  2. model.safetensors +1 -1
  3. runs/May20_21-47-51_e9c6c5e07544/events.out.tfevents.1716241721.e9c6c5e07544.5695.0 +2 -2
README.md CHANGED
@@ -17,12 +17,12 @@ model-index:
17
  name: ml-superb-subset
18
  type: ml-superb-subset
19
  config: ssw
20
- split: test
21
  args: ssw
22
  metrics:
23
  - name: Wer
24
  type: wer
25
- value: 0.8230088495575221
26
  ---
27
 
28
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -32,12 +32,22 @@ should probably proofread and complete it, then remove this comment. -->
32
 
33
  This model is a fine-tuned version of [facebook/wav2vec2-xls-r-300m](https://huggingface.co/facebook/wav2vec2-xls-r-300m) on the ml-superb-subset dataset.
34
  It achieves the following results on the evaluation set:
35
- - Loss: 1.2030
36
- - Wer: 0.8230
37
 
38
  ## Model description
39
 
40
- For CS224S. It's probably not very good.
 
 
 
 
 
 
 
 
 
 
41
 
42
  ### Training hyperparameters
43
 
@@ -51,23 +61,22 @@ The following hyperparameters were used during training:
51
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
52
  - lr_scheduler_type: linear
53
  - lr_scheduler_warmup_steps: 500
54
- - num_epochs: 50
55
  - mixed_precision_training: Native AMP
56
 
57
  ### Training results
58
 
59
- | Training Loss | Epoch | Step | Validation Loss | Wer |
60
- |:-------------:|:-----:|:----:|:---------------:|:------:|
61
- | 0.6745 | 5.0 | 50 | 1.0829 | 0.9204 |
62
- | 0.5247 | 10.0 | 100 | 1.0327 | 0.9204 |
63
- | 0.4094 | 15.0 | 150 | 0.9616 | 0.8938 |
64
- | 0.3286 | 20.0 | 200 | 0.9491 | 0.8761 |
65
- | 0.2973 | 25.0 | 250 | 0.9839 | 0.9381 |
66
- | 0.2678 | 30.0 | 300 | 1.0691 | 0.8761 |
67
- | 0.2701 | 35.0 | 350 | 1.0993 | 0.8496 |
68
- | 0.254 | 40.0 | 400 | 0.9991 | 0.8496 |
69
- | 0.2394 | 45.0 | 450 | 1.1706 | 0.8584 |
70
- | 0.2177 | 50.0 | 500 | 1.2030 | 0.8230 |
71
 
72
 
73
  ### Framework versions
 
17
  name: ml-superb-subset
18
  type: ml-superb-subset
19
  config: ssw
20
+ split: dev
21
  args: ssw
22
  metrics:
23
  - name: Wer
24
  type: wer
25
+ value: 1.0
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/wav2vec2-xls-r-300m](https://huggingface.co/facebook/wav2vec2-xls-r-300m) on the ml-superb-subset dataset.
34
  It achieves the following results on the evaluation set:
35
+ - Loss: 2.9777
36
+ - Wer: 1.0
37
 
38
  ## Model description
39
 
40
+ More information needed
41
+
42
+ ## Intended uses & limitations
43
+
44
+ More information needed
45
+
46
+ ## Training and evaluation data
47
+
48
+ More information needed
49
+
50
+ ## Training procedure
51
 
52
  ### Training hyperparameters
53
 
 
61
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
62
  - lr_scheduler_type: linear
63
  - lr_scheduler_warmup_steps: 500
64
+ - num_epochs: 10
65
  - mixed_precision_training: Native AMP
66
 
67
  ### Training results
68
 
69
+ | Training Loss | Epoch | Step | Validation Loss | Wer |
70
+ |:-------------:|:------:|:----:|:---------------:|:---:|
71
+ | 11.645 | 1.0471 | 100 | 10.4228 | 1.0 |
72
+ | 3.2477 | 2.0942 | 200 | 3.1142 | 1.0 |
73
+ | 3.1938 | 3.1414 | 300 | 3.0785 | 1.0 |
74
+ | 3.3024 | 4.1885 | 400 | 3.2026 | 1.0 |
75
+ | 3.1523 | 5.2356 | 500 | 3.1474 | 1.0 |
76
+ | 3.1189 | 6.2827 | 600 | 3.0652 | 1.0 |
77
+ | 3.0489 | 7.3298 | 700 | 3.0268 | 1.0 |
78
+ | 2.9999 | 8.3770 | 800 | 2.9902 | 1.0 |
79
+ | 2.9698 | 9.4241 | 900 | 2.9777 | 1.0 |
 
80
 
81
 
82
  ### Framework versions
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:98a1f8116efd58032767cf98581919ffbf7bfddf18ebb188a46346539759fe87
3
  size 1261938680
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:47b55480b924b335464fdb238d6f0519a3bd9f24c0e73603afb85057463163ce
3
  size 1261938680
runs/May20_21-47-51_e9c6c5e07544/events.out.tfevents.1716241721.e9c6c5e07544.5695.0 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:12008b04338332e0becf68348746961e9a18742e3a0fce81732ffc878c11e523
3
- size 12788
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f8e56f4e870d649571fae792f64fc3aefb820c753120aa471718bc67da8a73aa
3
+ size 13353