Searching for songs just got smoother.
Google announced last week that they will begin testing a brand new feature that will allow Android users to search for songs on YouTube simply by humming or singing.
“We’re experimenting with the ability for folks to search for a song on YouTube by humming or recording a song that’s currently being played,” reads a statement on YouTube’s support page.
According to the page, those who are in the beta group can toggle the new feature from YouTube’s voice search and “hum or record the song you’re searching for 3 or more seconds.”
Once the song has been identified by the software, the video-sharing website will send the user all “relevant official music content, user-generated videos, and/or Shorts featuring the searched song in the YouTube app.”
The Google-owned company stated that while only a small portion of Android users currently have the feature, the brand-new update will soon be released to everyone.
Should the update prove successful, the new feature will put YouTube in direct competition with the Apple-owned app Shazam.
According to Google, their goal is two-fold. One goal is to “make it easier for users to find” and engage with content on YouTube while the other is to relieve some of the pressure put on creators to upload multiple times a day.
The Post reached out to Google for comment.
This is not the first time Google has integrated sound-to-search technology into one of its platforms.
In 2020, the tech giant reportedly launched the feature on the Google app, Google Search widget and Google Assistant.
However, the main difference is that users must hum the tune for 10 to 15 seconds rather than three.
“When you hum a melody into Search, our machine learning models transform the audio into a number-based sequence representing the song’s melody,” explained the search engine. “Our models are trained to identify songs based on a variety of sources, including humans singing, whistling or humming, as well as studio recordings.”
“The algorithms also take away all the other details, like accompanying instruments and the voice’s timbre and tone. What we’re left with is the song’s number-based sequence, or the fingerprint,” continued the statement.
“We compare these sequences to thousands of songs from around the world and identify potential matches in real time.”
Source
