Hugging Face's logo

herwoww
/
mms-MGB5

Automatic Speech Recognition
Transformers
PyTorch
wav2vec2
Generated from Trainer
Inference Endpoints
Model card Files Community
herwoww commited on
Commit
fa9c4a9
•
1 Parent(s): 6a87c34

Model save

Browse files
Files changed (3) hide show
  1. README.md +84 -0
  2. adapter.ara.safetensors +3 -0
  3. pytorch_model.bin +1 -1
README.md ADDED
@@ -0,0 +1,84 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: cc-by-nc-4.0
3
+ base_model: facebook/mms-1b-all
4
+ tags:
5
+ - generated_from_trainer
6
+ metrics:
7
+ - wer
8
+ model-index:
9
+ - name: mms-MGB5
10
+ results: []
11
+ ---
12
+
13
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
14
+ should probably proofread and complete it, then remove this comment. -->
15
+
16
+ # mms-MGB5
17
+
18
+ This model is a fine-tuned version of [facebook/mms-1b-all](https://huggingface.co/facebook/mms-1b-all) on the None dataset.
19
+ It achieves the following results on the evaluation set:
20
+ - Loss: 1.3392
21
+ - Wer: 0.9101
22
+
23
+ ## Model description
24
+
25
+ More information needed
26
+
27
+ ## Intended uses & limitations
28
+
29
+ More information needed
30
+
31
+ ## Training and evaluation data
32
+
33
+ More information needed
34
+
35
+ ## Training procedure
36
+
37
+ ### Training hyperparameters
38
+
39
+ The following hyperparameters were used during training:
40
+ - learning_rate: 1e-05
41
+ - train_batch_size: 14
42
+ - eval_batch_size: 8
43
+ - seed: 42
44
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
45
+ - lr_scheduler_type: constant_with_warmup
46
+ - lr_scheduler_warmup_steps: 50
47
+ - num_epochs: 4
48
+
49
+ ### Training results
50
+
51
+ | Training Loss | Epoch | Step | Validation Loss | Wer |
52
+ |:-------------:|:-----:|:----:|:---------------:|:------:|
53
+ | 7.0257 | 0.16 | 250 | 6.2113 | 1.0041 |
54
+ | 3.2731 | 0.33 | 500 | 3.2218 | 1.0 |
55
+ | 2.8821 | 0.49 | 750 | 2.8522 | 1.0 |
56
+ | 2.1862 | 0.66 | 1000 | 2.0623 | 0.9721 |
57
+ | 1.6714 | 0.82 | 1250 | 1.6496 | 0.9398 |
58
+ | 1.7441 | 0.99 | 1500 | 1.5210 | 0.9314 |
59
+ | 1.4705 | 1.15 | 1750 | 1.4589 | 0.9251 |
60
+ | 1.4794 | 1.32 | 2000 | 1.4186 | 0.9213 |
61
+ | 1.4261 | 1.48 | 2250 | 1.3921 | 0.9175 |
62
+ | 1.3905 | 1.64 | 2500 | 1.3732 | 0.9185 |
63
+ | 1.4405 | 1.81 | 2750 | 1.3548 | 0.9134 |
64
+ | 1.3561 | 1.97 | 3000 | 1.3398 | 0.9093 |
65
+ | 1.4154 | 2.14 | 3250 | 1.3280 | 0.9128 |
66
+ | 1.361 | 2.3 | 3500 | 1.3161 | 0.9082 |
67
+ | 1.3421 | 2.47 | 3750 | 1.3064 | 0.9037 |
68
+ | 1.3615 | 2.63 | 4000 | 1.2976 | 0.9005 |
69
+ | 1.3517 | 2.8 | 4250 | 1.2996 | 0.9056 |
70
+ | 1.4316 | 2.96 | 4500 | 1.3136 | 0.9091 |
71
+ | 1.3946 | 3.12 | 4750 | 1.3534 | 0.9143 |
72
+ | 1.4888 | 3.29 | 5000 | 1.3797 | 0.9242 |
73
+ | 1.4223 | 3.45 | 5250 | 1.3707 | 0.9289 |
74
+ | 1.4616 | 3.62 | 5500 | 1.3463 | 0.9205 |
75
+ | 1.4196 | 3.78 | 5750 | 1.3362 | 0.9158 |
76
+ | 1.4163 | 3.95 | 6000 | 1.3392 | 0.9101 |
77
+
78
+
79
+ ### Framework versions
80
+
81
+ - Transformers 4.33.2
82
+ - Pytorch 2.0.1
83
+ - Datasets 2.19.1
84
+ - Tokenizers 0.13.3
adapter.ara.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:675d6af37ec48c2d8c2f8ffe565e8ecc30778f08282084a272b00311669a2ac4
3
+ size 8875400
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:e3995c609eb1d034162917c5ad9c349c4ba465472351fc0a38f310fd95f71247
3
  size 3859211405
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ed3210bd8ce5aa36918a5476be3078c740d12c209c787373ff2f52f2e5c33289
3
  size 3859211405