Cahya Wirawan
added first commit 39212b7
1
---
2
language: tr
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: Wav2Vec2 Base Turkish by Cahya
15
  results:
16
  - task: 
17
      name: Speech Recognition
18
      type: automatic-speech-recognition
19
    dataset:
20
      name: Common Voice tr
21
      type: common_voice
22
      args: tr
23
    metrics:
24
       - name: Test WER
25
         type: wer
26
         value: 13.70
27
---
28
29
# Wav2Vec2-Large-XLSR-Turkish
30
31
This is the model for Wav2Vec2-Base-Turkish-Artificial-CV, a fine-tuned 
32
[cahya/wav2vec2-base-turkish-artificial](https://huggingface.co/cahya/wav2vec2-base-turkish-artificial)
33
model on [Turkish Common Voice dataset](https://huggingface.co/datasets/common_voice).
34
35
When using this model, make sure that your speech input is sampled at 16kHz.
36
37
## Usage
38
The model can be used directly (without a language model) as follows:
39
```python
40
import torch
41
import torchaudio
42
from datasets import load_dataset
43
from transformers import Wav2Vec2ForCTC, Wav2Vec2Processor
44
45
test_dataset = load_dataset("common_voice", "tr", split="test[:2%]")
46
47
processor = Wav2Vec2Processor.from_pretrained("cahya/wav2vec2-base-turkish-artificial-cv")
48
model = Wav2Vec2ForCTC.from_pretrained("cahya/wav2vec2-base-turkish-artificial-cv")
49
50
51
# Preprocessing the datasets.
52
# We need to read the aduio files as arrays
53
def speech_file_to_array_fn(batch):
54
  speech_array, sampling_rate = torchaudio.load(batch["path"])
55
  resampler = torchaudio.transforms.Resample(sampling_rate, 16_000)
56
  batch["speech"] = resampler(speech_array).squeeze().numpy()
57
  return batch
58
59
test_dataset = test_dataset.map(speech_file_to_array_fn)
60
inputs = processor(test_dataset[:2]["speech"], sampling_rate=16_000, return_tensors="pt", padding=True)
61
62
with torch.no_grad():
63
  logits = model(inputs.input_values, attention_mask=inputs.attention_mask).logits
64
65
predicted_ids = torch.argmax(logits, dim=-1)
66
67
print("Prediction:", processor.batch_decode(predicted_ids))
68
print("Reference:", test_dataset[:2]["sentence"])
69
```
70
71
72
## Evaluation
73
74
The model can be evaluated as follows on the Turkish test data of Common Voice.
75
76
```python
77
import torch
78
import torchaudio
79
from datasets import load_dataset, load_metric
80
from transformers import Wav2Vec2ForCTC, Wav2Vec2Processor
81
import re
82
83
test_dataset = load_dataset("common_voice", "tr", split="test")
84
wer = load_metric("wer")
85
86
processor = Wav2Vec2Processor.from_pretrained("cahya/wav2vec2-base-turkish-artificial-cv")
87
model = Wav2Vec2ForCTC.from_pretrained("cahya/wav2vec2-base-turkish-artificial-cv") 
88
model.to("cuda")
89
90
chars_to_ignore_regex = '[\,\?\.\!\-\;\:\"\“\‘\”\'\`…\’»«]'
91
92
# Preprocessing the datasets.
93
# We need to read the aduio files as arrays
94
def speech_file_to_array_fn(batch):
95
  batch["sentence"] = re.sub(chars_to_ignore_regex, '', batch["sentence"]).lower()
96
  speech_array, sampling_rate = torchaudio.load(batch["path"])
97
  resampler = torchaudio.transforms.Resample(sampling_rate, 16_000)
98
  batch["speech"] = resampler(speech_array).squeeze().numpy()
99
  return batch
100
101
test_dataset = test_dataset.map(speech_file_to_array_fn)
102
103
# Preprocessing the datasets.
104
# We need to read the aduio files as arrays
105
def evaluate(batch):
106
  inputs = processor(batch["speech"], sampling_rate=16_000, return_tensors="pt", padding=True)
107
108
  with torch.no_grad():
109
    logits = model(inputs.input_values.to("cuda")).logits
110
111
  pred_ids = torch.argmax(logits, dim=-1)
112
  batch["pred_strings"] = processor.batch_decode(pred_ids)
113
  return batch
114
115
result = test_dataset.map(evaluate, batched=True, batch_size=8)
116
117
print("WER: {:2f}".format(100 * wer.compute(predictions=result["pred_strings"], references=result["sentence"])))
118
```
119
120
**Test Result**: 13.70 %
121
122
## Training
123
124
The Common Voice `train`, `validation`, other and invalidated 
125
126
The script used for training can be found [here](https://github.com/cahya-wirawan/indonesian-speech-recognition) 
127