SopiMlab

DeepLearningWithAudio

GANSpaceSynth

The hybrid architecture is based on the idea that we could simply apply a GAN method (GANSpace) to another GAN model (GANSynth).

py

py/pyext – Python scripting objects for Pure Data and Max

DeepLearningWithAudio18

This repository contains the deep learning with audio examples and the course materials for DOM-E5129 - Intelligent Computational Media. Documentation on different Deep Learning Audio systems as well as instructions on using some of them. Tools for loading, playing and plotting audio. Some working simple classifiers Non-working sample-level/raw audio GANs Python scripts for sorting different popular datasets

PESI_OnBody

PESI mobile phone instrument, running custom software built with Objective-C and Libpd in iPhones

kinectTrackerRealTime

kinect trackerRealTime module is built using the OpenNI library with Processing and openFrameworks. It is fully scalable to any number of Kinect v1 and to track any number of users.

PESI_InSpace

The in-space component implements a Kinect motion tracking module that constantly computes the spatial position of the three performers and extracts information about their relations, such as relative distances, velocity, acceleration and alignment

SoundingBrush

CCREAIM

CCREAIM is a hybrid transformer architecture, utilising attention weights to provide the musician with potentially interesting information about the motivation behind the AI's decisions.

flext

flext – C++ development layer for Pure Data and Max

kinectDataServer

Streams the point-cloud information from the Kinect devices

kinectPlySaver

Determines the point-cloud information of each Kinect for calibration

ddsp

DDSP: Differentiable Digital Signal Processing

Document-Formats

Everything you need to know to publish using LaTeX or Word and then some.

flower

InSpace_with_Otherness

magenta

Magenta: Music and Art Generation with Machine Intelligence

MediaAI

Aalto University's Intelligent Computational Media course

NIME-bibliography

BibTeX files with information about all publications from the annual International Conference on New Interfaces for Musical Expression (NIME)

NOISA

NOISA project provides an intelligent system that acts to maintain and deepen the performers en- gagement by learning from the performer’s actions and behavioural patterns while playing NOISA instruments.

open-nsynth-super

Open NSynth Super is an experimental physical interface for the NSynth algorithm

prism-samplernn

Neural sound synthesis with TensorFlow 2

RAVE

Official implementation of the RAVE model: a Realtime Audio Variational autoEncoder

SopiLib

Video_Tutorial_Kinect_Tracker

Video Tutorial for integrating Kinect Tracking modules in PESI project; kinectPlySaver; kinectDataServer; kinectTrackerRealTime