Audio Experiments

Why does this repository exist?

This repository is a collection of experiments with audio. The goal is to learn more about audio processing and architectures that can be used for generating audio.

What does this repository contain?

1. HiFi-GAN

This is an implementation of the HiFi-GAN model in PyTorch lightning. Contains code for training it on LibriSpeech dataset.

There will be more models and experiments added in the future...

What is the purpose of this repository?

This is my playground for experimenting with audio/music. I will be adding more models and experiments as I learn more about other architectures and techniques. The final goal is to add multimodality in Smol-LM

Notes

• The code is not optimized for performance. It is written in a way that is easy for me to experiment with.
• Improvements and suggestions are welcome. Feel free to open an issue or a pull request.
• Experiments with Images are done at ImageExpts