File size: 2,659 Bytes
f2f25c6
1cffeed
 
c873842
 
5945eec
 
 
 
 
 
becf0de
df9f71a
 
1cffeed
c873842
df9f71a
 
 
 
c873842
 
df9f71a
c873842
 
df9f71a
c873842
 
 
 
 
 
 
 
 
 
f2f25c6
 
 
 
 
 
 
 
 
 
 
 
 
becf0de
f2f25c6
 
becf0de
f2f25c6
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1cffeed
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
---
language:
- ur
license: apache-2.0
library_name: transformers
tags:
- automatic-speech-recognition
- robust-speech-event
- hf-asr-leaderboard
datasets:
- mozilla-foundation/common_voice_8_0
metrics:
- wer
- cer
pipeline_tag: automatic-speech-recognition
base_model: Harveenchadha/vakyansh-wav2vec2-urdu-urm-60
model-index:
- name: wav2vec2-urdu
  results:
  - task:
      type: automatic-speech-recognition
      name: Urdu Speech Recognition
    dataset:
      name: Common Voice (Urdu)
      type: common_voice
    metrics:
    - type: wer
      value: 57.47
      name: WER
      config: load_metric("wer")
      verified: true
    - type: cer
      value: 32.68
      name: CER
      config: load_metric("cer")
      verified: true
---

<!-- 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. -->

# wav2vec2-large-xls-r-300m-Urdu

This model is a fine-tuned version of [Harveenchadha/vakyansh-wav2vec2-urdu-urm-60](https://huggingface.co/Harveenchadha/vakyansh-wav2vec2-urdu-urm-60) on the common_voice dataset.
It achieves the following results on the evaluation set:
- Wer: 0.5747
- Cer: 0.3268

## Model description
The training and valid dataset is 0.58 hours. It was hard to train any model on lower number of so I decided to take vakyansh-wav2vec2-urdu-urm-60 checkpoint and finetune the wav2vec2 model.  

## Training procedure
Trained on Harveenchadha/vakyansh-wav2vec2-urdu-urm-60 due to lesser number of samples.

### Training hyperparameters

The following hyperparameters were used during training:
- learning_rate: 0.0003
- train_batch_size: 64
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 128
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 100
- num_epochs: 100
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch | Step | Validation Loss | Wer    | Cer    |
|:-------------:|:-----:|:----:|:---------------:|:------:|:------:|
| 4.3054        | 16.67 | 50   | 9.0055          | 0.8306 | 0.4869 |
| 2.0629        | 33.33 | 100  | 9.5849          | 0.6061 | 0.3414 |
| 0.8966        | 50.0  | 150  | 4.8686          | 0.6052 | 0.3426 |
| 0.4197        | 66.67 | 200  | 12.3261         | 0.5817 | 0.3370 |
| 0.294         | 83.33 | 250  | 11.9653         | 0.5712 | 0.3328 |
| 0.2329        | 100.0 | 300  | 7.6846          | 0.5747 | 0.3268 |


### Framework versions

- Transformers 4.16.0.dev0
- Pytorch 1.10.1+cu102
- Datasets 1.17.1.dev0
- Tokenizers 0.11.0