File size: 3,695 Bytes
e633a83
 
a5521c2
e633a83
 
 
 
e640432
a5521c2
 
 
 
 
e633a83
 
 
245a7f9
e633a83
a5521c2
e633a83
 
 
78d9d91
e633a83
a5521c2
245a7f9
a5521c2
e633a83
 
bf41557
e633a83
78d9d91
bf41557
e633a83
bf41557
e633a83
 
bf41557
e633a83
 
 
 
 
bf41557
78d9d91
bf41557
e633a83
 
bf41557
e633a83
bf41557
e633a83
 
 
 
 
 
bf41557
e633a83
 
bf41557
e633a83
 
 
bf41557
e633a83
 
 
bf41557
e633a83
bf41557
a8be5f0
bf41557
e633a83
 
 
 
 
 
bf41557
e633a83
 
bf41557
e633a83
 
 
bf41557
e633a83
 
bf41557
e633a83
 
 
 
 
 
 
bf41557
e633a83
bf41557
e633a83
a8be5f0
e633a83
 
 
 
 
 
 
bf41557
e633a83
 
 
bf41557
e633a83
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
---
language: lv
license: apache-2.0
tags:
- audio
- automatic-speech-recognition
- speech
- xlsr-fine-tuning-week
datasets:
- common_voice
metrics:
- wer
base_model: facebook/wav2vec2-large-xlsr-53
model-index:
- name: jimregan/wav2vec2-large-xlsr-latvian-cv
  results:
  - task:
      type: automatic-speech-recognition
      name: Speech Recognition
    dataset:
      name: Common Voice lv
      type: common_voice
      args: lv
    metrics:
    - type: wer
      value: 29.95
      name: Test WER
---
# Wav2Vec2-Large-XLSR-Latvian

Fine-tuned [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53)
on the [Latvian Common Voice dataset](https://huggingface.co/datasets/common_voice).

When using this model, make sure that your speech input is sampled at 16kHz.

## Usage
The model can be used directly (without a language model) as follows:

```python
import torch
import torchaudio
from datasets import load_dataset
from transformers import Wav2Vec2ForCTC, Wav2Vec2Processor

test_dataset = load_dataset("common_voice", "lv", split="test[:2%]")

processor = Wav2Vec2Processor.from_pretrained("jimregan/wav2vec2-large-xlsr-latvian-cv")
model = Wav2Vec2ForCTC.from_pretrained("jimregan/wav2vec2-large-xlsr-latvian-cv")

resampler = torchaudio.transforms.Resample(48_000, 16_000)

# Preprocessing the datasets.
# We need to read the aduio files as arrays
def speech_file_to_array_fn(batch):
	speech_array, sampling_rate = torchaudio.load(batch["path"])
	batch["speech"] = resampler(speech_array).squeeze().numpy()
	return batch
	
test_dataset = test_dataset.map(speech_file_to_array_fn)
inputs = processor(test_dataset["speech"][:2], sampling_rate=16_000, return_tensors="pt", padding=True)

with torch.no_grad():
	logits = model(inputs.input_values, attention_mask=inputs.attention_mask).logits
predicted_ids = torch.argmax(logits, dim=-1)

print("Prediction:", processor.batch_decode(predicted_ids))
print("Reference:", test_dataset["sentence"][:2])
```

## Evaluation

The model can be evaluated as follows on the Latvian test data of Common Voice.

```python
import torch
import torchaudio
from datasets import load_dataset, load_metric
from transformers import Wav2Vec2ForCTC, Wav2Vec2Processor
import re

test_dataset = load_dataset("common_voice", "lv", split="test")
wer = load_metric("wer")

processor = Wav2Vec2Processor.from_pretrained("jimregan/wav2vec2-large-xlsr-latvian-cv")
model = Wav2Vec2ForCTC.from_pretrained("jimregan/wav2vec2-large-xlsr-latvian-cv") 
model.to("cuda")

chars_to_ignore_regex = '[,\?\.\!\;\:\"\“\%\‘\”\(\)\*\…\—\–\']'
resampler = torchaudio.transforms.Resample(48_000, 16_000)

# Preprocessing the datasets.
# We need to read the aduio files as arrays
def speech_file_to_array_fn(batch):
	batch["sentence"] = re.sub(chars_to_ignore_regex, '', batch["sentence"]).lower()
	speech_array, sampling_rate = torchaudio.load(batch["path"])
	batch["speech"] = resampler(speech_array).squeeze().numpy()
	return batch
	
test_dataset = test_dataset.map(speech_file_to_array_fn)

# Preprocessing the datasets.
# We need to read the audio files as arrays
def evaluate(batch):
	inputs = processor(batch["speech"], sampling_rate=16_000, return_tensors="pt", padding=True)
	with torch.no_grad():
		logits = model(inputs.input_values.to("cuda"), attention_mask=inputs.attention_mask.to("cuda")).logits	
	pred_ids = torch.argmax(logits, dim=-1)
	batch["pred_strings"] = processor.batch_decode(pred_ids)
	return batch
	
result = test_dataset.map(evaluate, batched=True, batch_size=8)
print("WER: {:2f}".format(100 * wer.compute(predictions=result["pred_strings"], references=result["sentence"])))
```

**Test Result**: 29.95 %