There is nothing more frustrating than being unable to remember the name of a song or any of the words but the tune is stuck in your head. Google has recognised this issue and is using machine learning technology to help get rid of that annoying earworm.
Google has added a new “hum to search” feature on Google Search which enables its search engine to scour the internet for a particular song based off of nothing more than a user humming, whistling, or singing. Google’s software will then analyse the audio you have provided and serve up either a single song, or range of songs it thinks matches your sample.
To use the new feature on your mobile device, open the latest version of the Google app or find your Google Search widget, tap the mic icon and say “what’s this song?” or click the “Search a song” button. Then start humming for 10-15 seconds. This feature is currently available in English on iOS, and in more than 20 languages on Android, with the company looking to expand on that offering in the future.
According to Google, when an individual hums a melody into the new search feature, its machine learning models transform the audio into a number-based sequence representing the song’s melody. The models have been trained to identify songs based on a number of sources, including humans singing, whistling or humming, as well as studio recordings.
Take a look at a demonstration of the few search feature below