Polymath is a music production tool that uses machine learning to convert any music library into a sample library. It automatically separates songs into stems, quantizes them to the same tempo and beat grid, analyzes musical structure and tonality, and converts audio to MIDI. The result is a searchable sample library that streamlines workflow for music producers, DJs, and ML audio developers.

Main Features:

1. Music source separation: Polymath uses the Demucs neural network to separate songs into stems, such as beats, bass, and vocals.

2. Musical structure segmentation/labeling: The sf_segmenter neural network is used to analyze the musical structure of songs, identifying elements such as verses and choruses.

3. Music Pitch Tracking and Key Detection: Polymath uses the Crepe neural network to track the pitch of music and detect the key it is in.

Use case:

• Music composition: Polymath makes it easy to combine elements from different songs to create unique compositions. Users can easily grab a beat from one track, a bassline from another, and install horns from a third, integrating them seamlessly into their digital audio workstation (DAW).

• DJ Sets: With Polymath’s search capability, DJs can quickly discover related tracks and create polished hour-long DJ sets.

• ML Audio Development: Polymath simplifies the process of creating a large music dataset for training generative models by automatically converting audio files to MIDI and providing a searchable sample library.

Conclusion:

Polymath is a powerful tool for music producers, DJs and ML audio developers. Its machine learning capabilities allow users to convert music libraries into sample libraries, separate songs into stems, analyze musical structure and tonality, and convert audio to MIDI. With its ease of use and practical applications in music composition, DJ sets, and ML audio development, Polymath streamlines workflows and improves creativity in the music industry.