File size: 1,541 Bytes
346b014
 
 
 
9139449
346b014
 
 
 
 
9139449
346b014
 
 
 
 
 
3d63ba1
7932189
3d63ba1
 
 
 
346b014
833e23b
 
 
346b014
 
1db1a91
 
1ae7cce
 
 
 
69101a4
1ae7cce
 
 
346b014
 
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
---
license: apache-2.0
language:
- en
pipeline_tag: automatic-speech-recognition
datasets:
- LRS3

tags:
- Audio Visual to Text
- Automatic Speech Recognition
---

## Model Description

These are model weights originally provided by the authors of the paper [Learning Audio-Visual Speech Representation by Masked Multimodal Cluster Prediction](https://arxiv.org/pdf/2201.02184.pdf).

<figure>
  <img src="https://huggingface.co/vumichien/AV-HuBERT/resolve/main/HuBert.png" alt="Audio-visual HuBERT">
  <figcaption>Audio-visual HuBERT
  </figcaption>
</figure>

Video recordings of speech contain correlated audio and visual information, providing a strong signal for speech representation learning from the speaker’s lip
movements and the produced sound. 

Audio-Visual Hidden Unit BERT (AV-HuBERT), a self-supervised representation learning framework for audio-visual speech, which masks multi-stream video input and predicts automatically discovered and iteratively refined multimodal hidden units. AV-HuBERT
learns powerful audio-visual speech representation benefiting both lip-reading and automatic speech recognition.

The official code of this paper in [here](https://github.com/facebookresearch/av_hubert)

## Example

<figure>
  <img src="https://huggingface.co/vumichien/AV-HuBERT/resolve/main/lipreading.gif" alt="Audio-Visual Speech Recognition">
  <figcaption> Speech Recognition from visual lip movement
  </figcaption>
</figure>

## Datasets
The authors trained the model on lip-reading benchmark LRS3 datasets (433 hours).