Music Synthesis in Python

Dror Ayalon — Sat, 07 Oct 2017 00:00:00 +0000

Description

Music Synthesis in Python

Python has become a one-stop-shop for everything audio - from cutting edge digital signal processing packages to music synthesis and music composition packages. This plethora of audio related packages allows developers and musicians to build the most creative projects and experiences in a single programming language.

The goal of this talk to provide attendees with the knowledge of the fabulous audio analysis and music creation tools available in Python.

Talk Overview

The talk will start with a presentation of one of my recent projects, Luncz (pronounces “Luntz”), a tool that analyzes live playing and generates a bass line to accompany the musician. Luncz helps the musician to stay within the musical context of the original idea and leaves enough room for this idea to be developed. We will breakdown Luncz technical components and learn about the capabilities of the DSP packages that were used to build it and the Csound Python interface that was used to generate the playback.

Followed by the project presentation, we will briefly review a variety of available Python tools and packages for music creation, music analysis, and Python interfaces to the most used digital music synthesis engines. Among these tools are pyAudio, LibROSA, pyAudio, athenaCL, Cecilia, Pyo, Soundgrain, and Csound’s, Supercollider’s and ChucK’s Python APIs.

About The Speaker

Dror Ayalon is a maker, programmer, user experience designer, and a user interface designer, and was the Product Manager for a few very successful start-ups. Dror is currently a graduate student at NYU. As a lifelong music lover, he researches and develops innovative music creation tools, using music information retrieval (MIR) techniques, digital signal processing (DSP), and machine learning algorithms, that will allow musicians to compose music in a variety of new ways and formats.

Creative Music Applications in Python

Dror Ayalon — Fri, 05 Oct 2018 00:00:00 +0000

Description

### Creative Music Applications in Python We are lucky to live in times when Python is the go-to programming language for machine learning, neural networks research, web applications, as well as for developments of generative media and digital art. Using a wide variety of Python open-source packages and RNN models, the Python community is now becoming the center stage for experimental media projects and creative applications.

The goal of this talk is to introduce attendees to the increasing variety of packages and pre-trained machine learning models for audio analysis, music creation, and text generation, that is now available in Python!

### Talk Overview The talk will start with a presentation of Dror's recent award-winning project, [Soundscape](http://www.soundsca.pe) - An online platform, that allows music lovers to record loops, and sync them with music by other people around the world automatically.

We will breakdown Soundscape's building blocks and see how to use audio analysis Python packages, such as [LibROSA](https://librosa.github.io/librosa/index.html) and [madmom](http://madmom.readthedocs.io/en/latest/index.html) (Recurrent Neural Network). Furthermore, we will learn how to overcome the challenges of serving these type of applications using the [Django Web Framework](https://www.djangoproject.com/).

Then, we will talk about easy ways to build similar applications using Python packages, such as [AudioOwl](https://github.com/dodiku/audioowl) and [MixingBear](https://github.com/dodiku/MixingBear), which were created recently by Dror Ayalon.

![AudioOwl.py and MixingBear.py](https://s3.amazonaws.com/soundscape-images-dev/AudioOwl_MixingBear.png)

Following the project presentation, we will review a few other Python open-source projects and tools, such as Google Magenta's Recurrent Neural Networks for generative music, WaveGAN for generative raw audio, and textgenrnn for text generation.

We will wrap-up with a few interesting projects to burst our inspiration!

### About The Speaker [Dror Ayalon](https://www.drorayalon.com) is a software engineer, product manager, and interaction designer. He researches and develops innovative music creation tools, using music information retrieval (MIR) techniques, digital signal processing (DSP), and machine learning algorithms, that will allow musicians to compose music in a variety of new ways and formats. Dror recently received his master's degree from NYU and now works full time for the Google Creative Lab in NYC.

PyVideo.org - Dror Ayalon

Music Synthesis in Python

Description

Music Synthesis in Python

Talk Overview

About The Speaker

Creative Music Applications in Python

Description