YouTube tests music recognition feature like Shazam
YouTube is experimenting with a new feature that allows a subset of Android users to identify songs by humming, singing, or recording snippets of music. The feature, currently in its testing phase, aims to make the video-sharing platform an even more popular destination for music enthusiasts.
According to YouTube's support page, the search-by-song capability is accessible through the Android version of the YouTube app. Users who are part of the experiment can switch from YouTube's voice search to this new feature, then hum, sing, or record a song for at least three seconds. The platform then identifies the song and directs the user to relevant YouTube videos, which could include the official music video, user-generated content, or YouTube Shorts.
The feature is an extension of technology previously launched by Google, YouTube's parent company, in 2020. Google's version of the feature, available on the Google app, Google Search widget, and Google Assistant, also allows users to identify songs by humming, whistling, or singing. However, Google's version requires a longer input of 10-15 seconds to identify a song. Both features are built on machine learning models that match a user's hum to a song's 'fingerprint' or signature melody, the company confirmed to TechCrunch.
While YouTube's experiment is currently limited to a small group of Android users, its broader rollout could have significant implications. YouTube is already a major platform for music discovery and consumption, and this feature could make it even more central to the music ecosystem.
The move also places YouTube in direct competition with other music recognition apps like Shazam, which was acquired by Apple in 2018. Shazam identifies songs by listening to recorded music rather than interpreting hums or vocal renditions. Other apps like SoundHound and MusixMatch offer similar humming and singing recognition capabilities but have not gained as much traction as YouTube and Google in the music recognition space.