Ilya Borovik

SyMuPe: Affective and Controllable Symbolic Music Performance (ACM MM '25, Outstanding Paper Award)

A python package for handling modern staff notation of music

A convenient MIDI tokenizer for Deep Learning networks, with multiple encoding strategies

An introduction course on Speech Synthesis and Voice Cloning (Skoltech ISP'25)

ScorePerformer: Expressive Piano Performance Rendering with Fine-Grained Control (ISMIR 2023)

DeepHumor: Image-based Meme Generation using Deep Learning

Material for the paper "Automatic Quality Assessment of Transcribed Piano MIDI" (ISMIR 2025 LBD)

A storage for audio samples used in other repos

A simple TTS model developed for the Speech Synthesis and Voice Cloning course (Skoltech ISP'25)

Transfer Learning and DNN tuning on a MNIST dataset

Graph Database Engine

Vector Quantization, in Pytorch

Multi-Instrumental Neural Network for multi-track sequence generation

Implementation of AudioLM, a SOTA Language Modeling Approach to Audio Generation out of Google Research, in Pytorch

A simple but complete full-attention transformer with a set of promising experimental features from various papers

Financial Quest with Oleg, Tinkoff voice assistant

Audio watermarking based on the Singular Value Decomposition

Simple Taxi Ordering Web App written in Java EE

Flowtron is an auto-regressive flow-based generative network for text to speech synthesis with control over speech variation and style transfer

An implementation of local windowed attention for language modeling

Fully featured implementation of Routing Transformer

Mellotron: a multispeaker voice synthesis model based on Tacotron 2 GST that can make a voice emote and sing without emotive or singing training data

No description provided.

Personalized tasks & achievements based on customer retail transaction data

Lifestyle advices based on the transaction data

Reformer, the efficient Transformer, in Pytorch

Magenta: Music and Art Generation with Machine Intelligence

Simple analysis of attacks on RSA

MNIST dataset model in a Docker image

Application of Simulated Annealing to the Travelling Salesman problem