This project uses deep learning models to generate unique aircraft sounds from BBC's sound effects, showcasing generative AI's potential in sound synthesis
Description: This repository contains the code for a machine learning internship project focused on sound synthesis using deep generative models. The project aims to generate unique synthetic sound datasets for various aircraft sounds using GANs (Generative Adversarial Networks) and SampleRNN. The primary objective is to automatically learn the underlying patterns and regularities in the input sound data and generate new sound examples that closely resemble the original aircraft sounds.
training dataset is located here: https://sound-effects.bbcrewind.co.uk/search?cat=Aircraft
This project showcases the application of deep generative models in sound synthesis and can serve as a valuable resource for those interested in generative AI and audio generation. Feel free to explore the code, experiment with different hyperparameters, and contribute to the advancement of sound synthesis techniques.