Voice Processing

Our patented singing evaluation technology is able to assess singing proficiency with incredible depth. Our system compares a user-rendition of a song to the original singer. We use our state-of-the-art proprietary music analysis software to extract an exact high-resolution musical representation of the original singer's rendition of the song, which is used as a benchmark.

Commonly available vocal-suppression systems use center-channel cancellation techniques, which need the tracks to be in stereo mode. Further, such methods often suppress other instruments besides the voice. Our state-of-the art vocal suppression technology enables the creation of psuedo-karaoke music tracks for users to sing-along with. Our system is able to work on stereo and mono tracks, and removes only the lead-vocal melody without affecting any of the background accompaniment.

Our state-of-the-art mixing technology enables the creation of customized user-generated content. Starting from the basic mono-recording of the user-singing, we are able to offer a host of processing options (much like a virtual recording studio) to create a truly memorable cover version. Users are much more likely to share these covers with their friends rather than a vanilla mono-track.

Our high-level voice-processing technologies (singing rating, voice suppression, mixing with effects) rely on very solid, robust, low-level acoustic-feature extraction algorithms. These have been optimized to work in real-time i.e. as the user's voice is being recorded.