https://github.com/abdouaziz/wavautoencoder
WavAutoEncoder
https://github.com/abdouaziz/wavautoencoder
asr masked ssl
Last synced: about 2 months ago
JSON representation
WavAutoEncoder
- Host: GitHub
- URL: https://github.com/abdouaziz/wavautoencoder
- Owner: abdouaziz
- License: mit
- Created: 2022-11-24T21:37:10.000Z (over 2 years ago)
- Default Branch: main
- Last Pushed: 2023-05-08T10:33:18.000Z (about 2 years ago)
- Last Synced: 2025-01-30T00:32:12.292Z (4 months ago)
- Topics: asr, masked, ssl
- Language: Python
- Homepage:
- Size: 34.2 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
## Masked WavAutoEncoder
This repository is an implementation of masked wavautoencoder based on Transformer architecture for self supervised speech representation .
## Install
```bash
$ pip install wav-autoencoder
```
## Usage
Simple example for using the model
```python
import torch
from wav_autoencoder import WavAutoEncoderConfig , WavAutoEncoderModel
>> config = WavAutoEncoderConfig()
>> model = WavAutoEncoderModel(config)
>> wav_signal = torch.randn(2, 1, 16000)
>> outputs = model(wav_signal)
>> print(outputs.shape)
```
## Todo
## Citations
```bibtex
@misc{
title = {WavAutoencoder: A Self-Supervised Framework for Learning Audio Representations},
author = {Abdou Aziz DIOP},
year = {2022}
}
```