The DJ Sessions


Spotify Patent Filing Signals New Karaoke Feature

| June 27, 2021

The patent details technology that will identify what song a user is singing in order to play back its instrumental version.

While digital streaming platforms have given us music to sing along with, it’s not often that the technology sings back. That might change with a patent recently filed by Spotify for a “Karaoke Query Processing System.” 

The news follows a September discovery by reverse engineer Jane Manchun Wong, who found an unreleased in-app feature that would allow users to adjust the vocal levels of songs. Paired with this patent request, it seems Spotify is making serious plans to extend their reach into karaoke, though the company has not publicly addressed its plans for the technology.

According to patent application, which was submitted on June 24th, this system is capable of matching audio clips of songs with their instrumental versions in order to play a “backing track,” Digital Music News reports.

“In some embodiments, a user starts singing a song a cappella in the presence of a karaoke system as described herein,” the patent details. “The karaoke system determines what the user is singing and begins playing an instrumental version of the determined song. In some embodiments, the karaoke system plays the instrumental version using the key and tempo at which the user is singing.”

Even more, the “Karaoke Query Processing System” has capabilities not unlike Snapchat‘s partnership with Shazam, which identifies songs playing near a user’s phone. 

Spotify’s patent reads: “There is a need for a karaoke system which is capable of determining a users’ desired song for a karaoke performance even if the user cannot remember the name of the song or any other attributes of the song except for the melody of the song…The matching process includes comparing at least a portion of the audio clip to a plurality of reference audio tracks and identifying, based on the comparing, the first portion of a particular reference track that corresponds to the audio sample.”


Written by


This post currently has no comments.

Leave a Reply