1
---
2
language: ta
3
datasets:
4
- common_voice
5
tags:
6
- audio
7
- automatic-speech-recognition
8
- speech
9
- xlsr-fine-tuning-week
10
license: apache-2.0
11
model-index:
12
- name: XLSR Wav2Vec2 Tamil by Manan Dey
13
  results:
14
  - task: 
15
      name: Speech Recognition
16
      type: automatic-speech-recognition
17
    dataset:
18
      name: Common Voice ta
19
      type: common_voice
20
      args: ta
21
    metrics:
22
       - name: Test WER
23
         type: wer
24
         value: 56.44
25
---
26
27
# Wav2Vec2-Large-XLSR-53-Tamil
28
29
Fine-tuned [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) in Tamil 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", "ta", split="test[:2%]").
43
44
processor = Wav2Vec2Processor.from_pretrained("manandey/wav2vec2-large-xlsr-tamil")
45
model = Wav2Vec2ForCTC.from_pretrained("manandey/wav2vec2-large-xlsr-tamil")
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 {language} 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", "ta", split="test")
82
wer = load_metric("wer")
83
84
processor = Wav2Vec2Processor.from_pretrained("manandey/wav2vec2-large-xlsr-tamil")
85
model = Wav2Vec2ForCTC.from_pretrained("manandey/wav2vec2-large-xlsr-tamil")
86
model.to("cuda")
87
88
chars_to_ignore_regex = '[\,\?\.\!\-\;\:\"\“\%\‘\”\�\’\–\(\)]'
89
resampler = torchaudio.transforms.Resample(48_000, 16_000)
90
91
# Preprocessing the datasets.
92
# We need to read the aduio files as arrays
93
def speech_file_to_array_fn(batch):
94
    batch["sentence"] = re.sub(chars_to_ignore_regex, '', batch["sentence"]).lower()
95
    speech_array, sampling_rate = torchaudio.load(batch["path"])
96
    batch["speech"] = resampler(speech_array).squeeze().numpy()
97
    return batch
98
99
test_dataset = test_dataset.map(speech_file_to_array_fn)
100
101
# Preprocessing the datasets.
102
# We need to read the aduio files as arrays
103
def evaluate(batch):
104
    inputs = processor(batch["speech"], sampling_rate=16_000, return_tensors="pt", padding=True)
105
106
    with torch.no_grad():
107
        logits = model(inputs.input_values.to("cuda"), attention_mask=inputs.attention_mask.to("cuda")).logits
108
109
    pred_ids = torch.argmax(logits, dim=-1)
110
    batch["pred_strings"] = processor.batch_decode(pred_ids)
111
    return batch
112
113
result = test_dataset.map(evaluate, batched=True, batch_size=8)
114
115
print("WER: {:2f}".format(100 * wer.compute(predictions=result["pred_strings"], references=result["sentence"])))
116
```
117
118
**Test Result**: 56.44%
119
120
121
## Training
122
123
The Common Voice `train`, `validation` datasets were used for training.
124