Google Unveils 'Hum to Search' To Find Those Earworm Songs Stuck In Your Head

ML song recognition
Have you ever had a song stuck in your head, and you do not quite know the lyrics, but you know the tune? These earworms can also be incredibly frustrating when you cannot find the song, but Google is now looking to solve this problem by allowing for “Hum to search” functionality.

Following an update to Google Search, “you can hum, whistle, or sing a melody to Google to solve your earworm.” All one must do is ask Google, “What’s this song?” and start humming for 10-15 seconds.
On iOS it works in English only, while it works for 20 plus languages on Android. According to Google, “After you’re finished humming, our machine learning algorithm helps identify a potential song,” with a percent accuracy rating. You can then select the match and get the song unstuck from your head.

This algorithm is an exciting and impressive use of machine learning. Google trained machine learning models by converting the hum’s tune into a number sequence, which can be compared to a tune number sequence database. As Google explains, “a song’s melody is like its fingerprint,” and when you compare the base fingerprint from the tune you hum and a whole list of songs, you can get a solid match.

This is undoubtedly a step above Shazaam-like services and can be incredibly powerful when you get a song stuck in your head. Furthermore, this is only the beginning of machine learning applications in day-to-day life. With this application of machine learning, though, “next time you can’t remember the name of some catchy song you heard on the radio or that classic jam your parents love, just start humming. You’ll have your answer in record time.”