keums/melodyExtraction_JDC

"Joint Detection and Classification of Singing Voice Melody Using Convolutional Recurrent Neural Networks"

46
/ 100
Emerging

This project helps music analysts and audio engineers automatically pinpoint and track the melody in recorded songs. You provide an audio file, and it outputs a list of timestamps paired with the fundamental frequency (pitch) of the singing voice at each moment. This is ideal for researchers studying vocal performance, music educators analyzing songs, or developers building tools for music information retrieval.

131 stars. No commits in the last 6 months.

Use this if you need to precisely extract the melodic pitch contour of the singing voice from an audio recording, distinguishing it from background music or other sounds.

Not ideal if you need to extract the melody from instrumental music, identify multiple concurrent melodic lines, or only need general voice activity detection without pitch information.

music-information-retrieval vocal-analysis audio-processing musicology sound-engineering
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 20 / 25

How are scores calculated?

Stars

131

Forks

27

Language

Jupyter Notebook

License

MIT

Last pushed

Dec 27, 2019

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/keums/melodyExtraction_JDC"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.