🍣 SaShiMi-796 Samples
Showcase of samples from our 🍣 SaShiMi re-implementation, a deep-learning model for audio generation. Developed for METU CENG 796 Deep Generative Models course in Spring 2023.
Recent Posts
Why use μ-law encoding?
A demonstration that shows the advantages of using μ-law encoding instead of linear encoding when working with audio data in machine learning.
Tackling the Generative Learning Trilemma with Denoising Diffusion GANs
A presentation I made about this ICLR 2022 Spotlight paper.
Hexxagon Devlog: WebRTC, UI
I updated my Hexxagon implementation. See previous devlog.
Serve a Cross-Origin Isolated Webpage using Python
Simple Python HTTP server for testing multi-threaded web applications.