File size: 3,928 Bytes
506fa99
c4e7470
ae64dce
f0e97b6
 
 
 
 
 
 
ae64dce
 
c4e7470
 
ae64dce
05d379e
f0e97b6
 
 
 
a27570e
 
 
f0e97b6
 
 
 
 
a27570e
f0e97b6
5b11ee7
 
 
 
a27570e
 
5b11ee7
 
 
 
 
a27570e
5b11ee7
 
 
 
a27570e
 
 
5b11ee7
 
 
 
 
a27570e
5b11ee7
f0e97b6
506fa99
 
ae64dce
 
 
 
 
05d379e
ae64dce
b3bb914
 
ae64dce
 
 
 
05d379e
 
ae64dce
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
b3bb914
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
ae64dce
 
 
 
 
 
 
f0e97b6
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
---
base_model: facebook/w2v-bert-2.0
license: mit
datasets:
- thennal/IMaSC
- vrclc/festvox-iiith-ml
- vrclc/openslr63
- thennal/msc
- mozilla-foundation/common_voice_16_1

metrics:
- wer
tags:
- generated_from_trainer
model-index:
- name: w2v2bert-Malayalam
  results:
  - task:
      type: automatic-speech-recognition
      name: Automatic Speech Recognition
    dataset:
      name: OpenSLR Malayalam -Test
      type: vrclc/openslr63
      config: ml
      split: test
      args: ml
    metrics:
    - type: wer
      value: 12.66
      name: WER
  - task:
      type: automatic-speech-recognition
      name: Automatic Speech Recognition
    dataset:
      name: Goole Fleurs
      type: google/fleurs
      config: ml
      split: test
      args: ml
    metrics:
    - type: wer
      value: 40.58
      name: WER
  - task:
      type: automatic-speech-recognition
      name: Automatic Speech Recognition
    dataset: 
      name: Common Voice 16 Malayalam
      type: mozilla-foundation/common_voice_16_1
      config: ml
      split: test
      args: ml
    metrics:
    - type: wer
      value: 52.10
      name: WER

---

<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->

# w2v-bert-2.0-nonstudio_and_studioRecords

This model is a fine-tuned version of [facebook/w2v-bert-2.0](https://huggingface.co/facebook/w2v-bert-2.0) on an these datasets: [IMASC](https://huggingface.co/datasets/thennal/IMaSC), [MSC](https://huggingface.co/datasets/smcproject/MSC), [OpenSLR Malayalam Train split](https://huggingface.co/datasets/vrclc/openslr63), [Festvox Malayalam](https://huggingface.co/datasets/vrclc/openslr63), [common_voice_16_1](https://huggingface.co/datasets/mozilla-foundation/common_voice_16_1) 
It achieves the following results on the evaluation set:
- Loss: 0.1722
- Wer: 0.1299


## Training procedure

Trained on NVIDIA A100 GPU

### Training hyperparameters

The following hyperparameters were used during training:
- learning_rate: 5e-05
- train_batch_size: 16
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- num_epochs: 10
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch | Step  | Validation Loss | Wer    |
|:-------------:|:-----:|:-----:|:---------------:|:------:|
| 1.1416        | 0.46  | 600   | 0.3393          | 0.4616 |
| 0.1734        | 0.92  | 1200  | 0.2414          | 0.3493 |
| 0.1254        | 1.38  | 1800  | 0.2205          | 0.2963 |
| 0.1097        | 1.84  | 2400  | 0.2157          | 0.3133 |
| 0.0923        | 2.3   | 3000  | 0.1854          | 0.2473 |
| 0.0792        | 2.76  | 3600  | 0.1939          | 0.2471 |
| 0.0696        | 3.22  | 4200  | 0.1720          | 0.2282 |
| 0.0589        | 3.68  | 4800  | 0.1768          | 0.2013 |
| 0.0552        | 4.14  | 5400  | 0.1635          | 0.1864 |
| 0.0437        | 4.6   | 6000  | 0.1501          | 0.1826 |
| 0.0408        | 5.06  | 6600  | 0.1500          | 0.1645 |
| 0.0314        | 5.52  | 7200  | 0.1559          | 0.1655 |
| 0.0317        | 5.98  | 7800  | 0.1448          | 0.1553 |
| 0.022         | 6.44  | 8400  | 0.1592          | 0.1590 |
| 0.0218        | 6.9   | 9000  | 0.1431          | 0.1458 |
| 0.0154        | 7.36  | 9600  | 0.1514          | 0.1366 |
| 0.0141        | 7.82  | 10200 | 0.1540          | 0.1383 |
| 0.0113        | 8.28  | 10800 | 0.1558          | 0.1391 |
| 0.0085        | 8.74  | 11400 | 0.1612          | 0.1356 |
| 0.0072        | 9.2   | 12000 | 0.1697          | 0.1289 |
| 0.0046        | 9.66  | 12600 | 0.1722          | 0.1299 |


### Framework versions

- Transformers 4.39.3
- Pytorch 2.1.1+cu121
- Datasets 2.16.1
- Tokenizers 0.15.1