Дмитрий Пахомов - Development of Drum Transcription Software Based on Convolutional Neural Networks

GitHub: Abstract: In my bachelor’s thesis, I present a method to generate drum notation from raw audio data using a Convolutional Neural Network (CNN) model. The main premise of the work is to set the foundations of a drum scoring software built on a personally curated dataset. The significance of my project is that it is among the first to attempt to take a trained Automatic Drum Transcription (ADT) neural network into an easily accessible software for drum players of any skill level. According to a survey conducted within an international drumming school network, this project provides the most merit to hobbyist-professional drum players, and especially drum instructors that spend large amounts of time creating high-quality drum transcriptions. In addition, I also propose a UX/UI mockup that could be used to create a fully functional automatic drum score app. If you are interested in my project or want to read the full thesis, do not hesitate to contact me! Email: skittree@ Telegram: Data Fest 2023: Трек “Computer Vision“: Наши соц.сети: Telegram: Вконтакте: