santi-pdp/segan_pytorch
A Generative Adversarial Network for speech enhancement and denoising implemented in PyTorch.
Not currently ranked — collecting fresh signals.
star history
This repository implements SEGAN (Speech Enhancement GAN) and its improved variant SEGAN+ for audio denoising and dewhispering. The model uses a generator network trained adversarially to enhance noisy speech utterances. It provides training scripts, pretrained generator weights, and audio samples demonstrating the denoising capabilities on speech recordings.
Frequently asked
- What is santi-pdp/segan_pytorch?
- A Generative Adversarial Network for speech enhancement and denoising implemented in PyTorch.
- Is segan_pytorch open source?
- Yes — santi-pdp/segan_pytorch is open source, released under the MIT license.
- What language is segan_pytorch written in?
- santi-pdp/segan_pytorch is primarily written in Python.
- How popular is segan_pytorch?
- santi-pdp/segan_pytorch has 409 stars on GitHub.
- Where can I find segan_pytorch?
- santi-pdp/segan_pytorch is on GitHub at https://github.com/santi-pdp/segan_pytorch.