We have developed an application that extracts vocal sound from mixed sound by letting AI learn the spectrum of vocal sound. Screen shot 1 shows the mixed sound spectrum, and screen shot 2 shows the vocal sound spectrum extracted from the mixed sound by this application. There are still issues to be solved in order to improve sound quality and processing speed, so we plan to continue development in the future.
This application uses mobile-ffmpeg to decode audio data. #mobile-ffmpeg©2018-2020 MobileFFmpeg LGPL v3.0