1
---
2
language: pa-IN
3
datasets:
4
- common_voice 
5
metrics:
6
- wer
7
tags:
8
- audio
9
- automatic-speech-recognition
10
- speech
11
- xlsr-fine-tuning-week
12
license: apache-2.0
13
model-index:
14
- name: danurahul/wav2vec2-large-xlsr-pa-IN
15
  results:
16
  - task: 
17
      name: Speech Recognition
18
      type: automatic-speech-recognition
19
    dataset:
20
      name: Common Voice pa-IN
21
      type: common_voice
22
      args: pa-IN
23
    metrics:
24
       - name: Test WER
25
         type: wer
26
         value: 54.86
27
---
28
# Wav2Vec2-Large-XLSR-53-Punjabi
29
Fine-tuned [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on Punjabi using the [Common Voice](https://huggingface.co/datasets/common_voice).
30
When using this model, make sure that your speech input is sampled at 16kHz.
31
32
## Usage
33
34
The model can be used directly (without a language model) as follows:
35
36
```python
37
import torch
38
import torchaudio
39
from datasets import load_dataset
40
from transformers import Wav2Vec2ForCTC, Wav2Vec2Processor
41
42
test_dataset = load_dataset("common_voice", "pa-IN", split="test[:2%]") 
43
44
processor = Wav2Vec2Processor.from_pretrained("danurahul/wav2vec2-large-xlsr-pa-IN") 
45
model = Wav2Vec2ForCTC.from_pretrained("danurahul/wav2vec2-large-xlsr-pa-IN") 
46
47
resampler = torchaudio.transforms.Resample(48_000, 16_000)
48
49
# Preprocessing the datasets.
50
# We need to read the aduio files as arrays
51
def speech_file_to_array_fn(batch):
52
  speech_array, sampling_rate = torchaudio.load(batch["path"])
53
  batch["speech"] = resampler(speech_array).squeeze().numpy()
54
  return batch
55
56
test_dataset = test_dataset.map(speech_file_to_array_fn)
57
inputs = processor(test_dataset["speech"][:2], sampling_rate=16_000, return_tensors="pt", padding=True)
58
59
with torch.no_grad():
60
  logits = model(inputs.input_values, attention_mask=inputs.attention_mask).logits
61
62
predicted_ids = torch.argmax(logits, dim=-1)
63
64
print("Prediction:", processor.batch_decode(predicted_ids))
65
print("Reference:", test_dataset["sentence"][:2])
66
```
67
68
69
## Evaluation
70
71
The model can be evaluated as follows on the Punjabi test data of Common Voice. 
72
73
74
```python
75
import torch
76
import torchaudio
77
from datasets import load_dataset, load_metric
78
from transformers import Wav2Vec2ForCTC, Wav2Vec2Processor
79
import re
80
81
test_dataset = load_dataset("common_voice", "pa-IN", split="test") 
82
83
wer = load_metric("wer")
84
85
processor = Wav2Vec2Processor.from_pretrained("danurahul/wav2vec2-large-xlsr-pa-IN") 
86
87
model = Wav2Vec2ForCTC.from_pretrained("danurahul/wav2vec2-large-xlsr-pa-IN") 
88
89
model.to("cuda")
90
91
chars_to_ignore_regex = '[\\\\\\\\\\\\\\\\,\\\\\\\\\\\\\\\\?\\\\\\\\\\\\\\\\.\\\\\\\\\\\\\\\\!\\\\\\\\\\\\\\\\-\\\\\\\\\\\\\\\\;\\\\\\\\\\\\\\\\:\\\\\\\\\\\\\\\\"\\\\\\\\\\\\\\\\“\\\\\\\\\\\\\\\\%\\\\\\\\\\\\\\\\‘\\\\\\\\\\\\\\\\”\\\\\\\\\\\\\\\\�]'  
92
resampler = torchaudio.transforms.Resample(48_000, 16_000)
93
94
# Preprocessing the datasets.
95
# We need to read the aduio files as arrays
96
def speech_file_to_array_fn(batch):
97
  batch["sentence"] = re.sub(chars_to_ignore_regex, '', batch["sentence"]).lower()
98
  speech_array, sampling_rate = torchaudio.load(batch["path"])
99
  batch["speech"] = resampler(speech_array).squeeze().numpy()
100
  return batch
101
102
test_dataset = test_dataset.map(speech_file_to_array_fn)
103
104
# Preprocessing the datasets.
105
# We need to read the aduio files as arrays
106
def evaluate(batch):
107
  inputs = processor(batch["speech"], sampling_rate=16_000, return_tensors="pt", padding=True)
108
109
  with torch.no_grad():
110
    logits = model(inputs.input_values.to("cuda"), attention_mask=inputs.attention_mask.to("cuda")).logits
111
112
  pred_ids = torch.argmax(logits, dim=-1)
113
  batch["pred_strings"] = processor.batch_decode(pred_ids)
114
  return batch
115
116
result = test_dataset.map(evaluate, batched=True, batch_size=8)
117
118
print("WER: {:2f}".format(100 * wer.compute(predictions=result["pred_strings"], references=result["sentence"])))
119
```
120
121
**Test Result**: 100 %  
122
123
124
## Training
125
126
The Common Voice `train`, `validation` was used for training as well as validation and testing  #
127
128
The script used for training can be found https://github.com/rahul-art/huggingface_wav2vec2_punjabi/blob/main/Fine_Tune_XLSR_Wav2Vec2_on_Punjabi_ASR_with_%F0%9F%A4%97_Transformers.ipynb