File size: 5,040 Bytes
8d88236
 
 
 
 
101761d
 
8833830
8d88236
8edc0e7
7717449
728745f
7717449
728745f
8833830
 
 
 
 
 
 
55bb623
8833830
55bb623
 
 
 
8833830
 
 
 
55bb623
 
 
 
 
 
 
 
 
 
 
 
 
 
8d88236
0b362f5
8d88236
0b362f5
8d88236
0b362f5
233a80e
 
0b362f5
8d88236
0b362f5
8d88236
0b362f5
8d88236
5dadc39
8d88236
0b362f5
8d88236
0b362f5
 
8d88236
0b362f5
5dadc39
0b362f5
 
6b154c5
8d88236
 
 
 
6b154c5
d79204c
8d88236
 
 
 
 
5e3d456
8d88236
 
 
 
 
 
6b154c5
266740b
 
671b8e7
ed727c2
 
 
 
 
 
 
 
 
 
 
 
266740b
ed727c2
266740b
6b154c5
ed727c2
266740b
 
 
d79204c
6b154c5
266740b
 
ed727c2
 
 
266740b
 
 
 
6b154c5
ed727c2
266740b
 
ed727c2
 
266740b
ed727c2
266740b
 
08efb96
 
266740b
 
f27b797
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
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
---
language: en
datasets:
- librispeech_asr
tags:
- audio
- automatic-speech-recognition
- hf-asr-leaderboard
license: apache-2.0
widget:
- example_title: Librispeech sample 1
  src: https://cdn-media.huggingface.co/speech_samples/sample1.flac
- example_title: Librispeech sample 2
  src: https://cdn-media.huggingface.co/speech_samples/sample2.flac
model-index:
- name: wav2vec2-base-960h
  results:
  - task:
      name: Automatic Speech Recognition
      type: automatic-speech-recognition
    dataset:
      name: LibriSpeech (clean)
      type: librispeech_asr
      config: clean
      split: test
      args: 
        language: en
    metrics:
    - name: Test WER
      type: wer
      value: 3.4
  - task:
      name: Automatic Speech Recognition
      type: automatic-speech-recognition
    dataset:
      name: LibriSpeech (other)
      type: librispeech_asr
      config: other
      split: test
      args: 
        language: en
    metrics:
    - name: Test WER
      type: wer
      value: 8.6
---

# Wav2Vec2-Base-960h

[Facebook's Wav2Vec2](https://ai.facebook.com/blog/wav2vec-20-learning-the-structure-of-speech-from-raw-audio/)

The base model pretrained and fine-tuned on 960 hours of Librispeech on 16kHz sampled speech audio. When using the model
make sure that your speech input is also sampled at 16Khz.

[Paper](https://arxiv.org/abs/2006.11477)

Authors: Alexei Baevski, Henry Zhou, Abdelrahman Mohamed, Michael Auli

**Abstract**

We show for the first time that learning powerful representations from speech audio alone followed by fine-tuning on transcribed speech can outperform the best semi-supervised methods while being conceptually simpler. wav2vec 2.0 masks the speech input in the latent space and solves a contrastive task defined over a quantization of the latent representations which are jointly learned. Experiments using all labeled data of Librispeech achieve 1.8/3.3 WER on the clean/other test sets. When lowering the amount of labeled data to one hour, wav2vec 2.0 outperforms the previous state of the art on the 100 hour subset while using 100 times less labeled data. Using just ten minutes of labeled data and pre-training on 53k hours of unlabeled data still achieves 4.8/8.2 WER. This demonstrates the feasibility of speech recognition with limited amounts of labeled data.

The original model can be found under https://github.com/pytorch/fairseq/tree/master/examples/wav2vec#wav2vec-20.


# Usage

To transcribe audio files the model can be used as a standalone acoustic model as follows:

```python
 from transformers import Wav2Vec2Processor, Wav2Vec2ForCTC
 from datasets import load_dataset
 import torch
 
 # load model and tokenizer
 processor = Wav2Vec2Processor.from_pretrained("facebook/wav2vec2-base-960h")
 model = Wav2Vec2ForCTC.from_pretrained("facebook/wav2vec2-base-960h")
     
 # load dummy dataset and read soundfiles
 ds = load_dataset("patrickvonplaten/librispeech_asr_dummy", "clean", split="validation")
 
 # tokenize
 input_values = processor(ds[0]["audio"]["array"], return_tensors="pt", padding="longest").input_values  # Batch size 1
 
 # retrieve logits
 logits = model(input_values).logits
 
 # take argmax and decode
 predicted_ids = torch.argmax(logits, dim=-1)
 transcription = processor.batch_decode(predicted_ids)
 ```
 
 ## Evaluation

First, ensure the required Python packages are installed. We'll require `transformers` for running the Wav2Vec2 model,
`datasets` for loading the LibriSpeech dataset, and `evaluate` plus `jiwer` for computing the word-error rate (WER):

```
pip install --upgrade pip
pip install --upgrade transformers datasets evaluate jiwer
```

The following code snippet shows how to evaluate **facebook/wav2vec2-base-960h** on LibriSpeech's "clean" and "other" test data.
The batch size can be set according to your device, and is set to `8` by default:

```python
import torch
from datasets import load_dataset
from transformers import Wav2Vec2ForCTC, Wav2Vec2Processor
from evaluate import load

librispeech_eval = load_dataset("librispeech_asr", "clean", split="test")

model = Wav2Vec2ForCTC.from_pretrained("facebook/wav2vec2-base-960h").to("cuda")
processor = Wav2Vec2Processor.from_pretrained("facebook/wav2vec2-base-960h")

def map_to_pred(batch):
    audios = [audio["array"] for audio in batch["audio"]]
    sampling_rate = batch["audio"][0]["sampling_rate"]
    input_values = processor(audios, sampling_rate=sampling_rate, return_tensors="pt", padding="longest").input_values
    with torch.no_grad():
        logits = model(input_values.to("cuda")).logits

    predicted_ids = torch.argmax(logits, dim=-1)
    transcription = processor.batch_decode(predicted_ids)
    batch["transcription"] = [t for t in transcription]
    return batch

result = librispeech_eval.map(map_to_pred, batched=True, batch_size=8, remove_columns=["audio"])
wer = load("wer")

print("WER:", wer.compute(references=result["text"], predictions=result["transcription"]))
```

*Result (WER)*:

| "clean" | "other" |
|---|---|
| 3.4 | 8.6 |