Vocal Extractor Overview

  • You can extract or remove vocal from your music in Music Library by just pressing a button on the main screen (*1)
  • Generated vocal and backing tracks can be saved to files and exported to  iCloud, Dropbox, Google Drive and etc., so that you can use outputs on the other applications (*2)
  • Although the quality of vocal separation is varied depend on the source data as other applications, we use Deep Learning technologies which allow us to separate vocal from monaural recordings or not center pan vocal recordings
  • Vocal Extractor also supports importing mp3,m4a,wav  music files from iCloud,Dropbox,Google Drive and etc. as source data. Therefore you can record live performance by other application, import those recordings to this app and you will be able to remove or extract vocal from your recordings
  • The embedded player has a feature to switch  between original, vocal, and backing track quickly while playing

*1. Only downloaded on the device and NOT DRM protected songs
can be chosen as source data

*2 .Saved output files will be sampling rate 16KHz m4a monaural format.
Saving file is an in-App purchase feature. So we recommend to
confirm the quality of vocal separation on the embedded player then
consider to purchase this feature

Vocal Extractor is developed based on the following research paper

Andreas Jansson, Eric J. Humphrey, Nicola Montecchio, Rachel Bittner, Aparna Kumar, Tillman Weyde, Singing Voice Separation with Deep U-Net Convolutional Networks, Proceedings of the 18th International Society for Music Information Retrieval Conference (ISMIR), 2017.

https://ismir2017.smcnus.org/wp-content/uploads/2017/10/171_Paper.pdf