https://github.com/npuichigo/waveglow
A PyTorch implementation of the WaveGlow: A Flow-based Generative Network for Speech Synthesis
https://github.com/npuichigo/waveglow
neural-vocoder text-to-speech waveglow
Last synced: 4 months ago
JSON representation
A PyTorch implementation of the WaveGlow: A Flow-based Generative Network for Speech Synthesis
- Host: GitHub
- URL: https://github.com/npuichigo/waveglow
- Owner: npuichigo
- License: apache-2.0
- Created: 2018-11-04T16:30:44.000Z (about 7 years ago)
- Default Branch: master
- Last Pushed: 2018-11-06T14:42:38.000Z (about 7 years ago)
- Last Synced: 2025-07-05T22:37:22.126Z (5 months ago)
- Topics: neural-vocoder, text-to-speech, waveglow
- Language: Python
- Size: 25.4 KB
- Stars: 205
- Watchers: 16
- Forks: 35
- Open Issues: 2
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
- Awesome-pytorch-list-CNVersion - torch_waveglow
- Awesome-pytorch-list - torch_waveglow - based Generative Network for Speech Synthesis. (Paper implementations / Other libraries:)
README
# WaveGlow
A PyTorch implementation of the [WaveGlow: A Flow-based Generative Network for Speech Synthesis](https://arxiv.org/abs/1811.00002)
## Quick Start:
1. Install requirements:
```
pip install -r requirements.txt
```
2. Download dataset:
```
wget http://festvox.org/cmu_arctic/cmu_arctic/packed/cmu_us_slt_arctic-0.95-release.tar.bz2
tar xf cmu_us_slt_arctic-0.95-release.tar.bz2
```
3. Extract features:
feature extracting pipeline is the same as [tacotron](https://github.com/keithito/tacotron)
4. Training with default hyperparams:
```
python train.py
```
5. Synthesize from model:
```
python generate.py --checkpoint=/path/to/model --local_condition_file=/path/to/local_conditon
```
## Notes:
* This is not official implementation, some details are not necessarily correct.
* Work in progress.