We’ve all been there. We can’t remember the name of a song or its lyrics, but we remember the general tune. It drives us all crazy that we may never know what the name of the song is or who sings it.

Enter Google.

The major tech company has created a solution to this first-world problem. Last week, they announced a “hum to search” feature in the Google app and Google Search Widget, which allows users to hum, whistle or sing a melody into the app, which then tries to find the song that the user has in mind.

To use this new feature, Google App or Google Search Widget users tap the mic icon on their mobile device and say “What’s this song?” Alternatively, they can click the “Search a song” button and start humming, whistling or singing the tune for 10-15 seconds.

View this post on Instagram

From breakthrough improvements on the quality of results to new ways to search, Google is here to help you understand the world’s information. Get the rundown of the latest Google Search updates at the link in bio. ✨

A post shared by Google (@google) on Oct 16, 2020 at 2:13pm PDT

//www.instagram.com/embed.js

After the user is finished humming, whistling or singing, Google’s artificial intelligence (AI) algorithms try to identify potential song matches and display the most likely options based on the tune. Users can then select the best match and explore information on the song and artist, find the lyrics, view any accompanying music videos or listen to the song on a music streaming app (such as Spotify or Apple Music), or even check out other recordings of the song if they’re available.

The “Hum To Search” feature is currently available in English on iOS operating systems, in more than 20 different languages on Android and through Google Assistant.

Krishna Kumar, the Senior Product Manager for Google Search, explained how the feature works in a blog post: “When you hum a melody into Search, our machine learning models transform the audio into a number-based sequence representing the song’s melody.”

The AI models are trained to identify songs based on a number of sources. The algorithms also remove other details of the songs, like accompanying instruments and the voice’s tone and timbre. “What we’re left with is the song’s number-based sequence or the fingerprint,” Kumar wrote.

You can watch a full video presentation of the “Hum to Search” feature below:

Featured Image: Google