Neural network Polymath
- Business model: Free
-
Go to website
-
Polymath is a tool that uses machine learning to transform any music library (like your hard drive or YouTube) into a sample library for music production.
It automatically divides songs into stems (beats, bass, etc.), quantizes them to the same tempo and grid (e.g. 120 beats per minute), analyzes musical structure (e.g. verse, chorus, etc.), tonality (eg C4, E3, etc.) and other information (timbre, volume, etc.), and also converts audio to MIDI. The result is a searchable sample library that simplifies the ML-based workflow for music producers, DJs, and audio developers.
-
Polymath: instructions
Examples of using
Polymath makes it easy to combine elements from different songs to create unique new compositions: simply take the beat from a Funkadelic track, the bass line from a Tito Puente piece, and the matching horns from a Fela Kuti song and seamlessly integrate them into your DAW in record time. Using Polymath's search function to find similar tracks, it's easy to create a polished, hour-long mash-up DJ set. For ML developers, Polymath simplifies the process of creating large music data sets, for training generative models, and more.
How it works?
- Music source separation is performed using the Demucs neural network
- Segmentation/labeling of musical structure is performed using the sf_segmenter neural network
- Music pitch tracking and key detection are done using Crepe neural network
- Transcription of music to MIDI is performed using the Basic Pitch neural network
- Quantization and equalization of music is done using pyrubberband
- Searching and processing of musical information is carried out using librosa
Polymath: how to use, instructions, video
The Polymath neural network was first published on 03/07/2023 21:51:52 and manually edited on 04/19/2024 22:33:28.
Blog