“Hum to Search” Let Google do the Job.

October 19, 2020

As technology evolves, many developers and tech-savvy people discover many ways of bringing comfort to everyone’s lives. One of these is Google.

Google became the one-stop searching location of people who wants instant answers, whether for information, DIY instructions, and many more. Also, Google is famous for introducing a wide array of kickass features to its app, whether for iOS or Android. While many still don’t value the importance of a particular tool and find it unnecessary, they are made to make life easier.

“Hum to Search”

 It is fair to say that anyone has already experience Last Song Syndrome. But worse is that you can’t recall the title of the song either the singer. Having it said, from now on, you can search for a piece that keeps playing in your head thru the help of “Hum to Search” using the Google App on both iOS and Android. To do this, open the Google app, press the microphone and speak “What’s this song” or click the “Search a Song” button.

How to Use the Hum to Search

To ensure that everything goes right, you need to hum for about 10 to 15 seconds before Google Search can perform its searching and give you all the possibilities listed results. It will provide you details such as all the match percentage and even the cover artist name on the result page. But wait, there is more. If you want to make it more easy, you can use instead of the Hum to Search Feature thru Google Assistant and say, “Hey Google, what is this song.”

Google said that currently, the feature is available in English for iOS, meanwhile more than 20 languages on Android. The company hopes to provide more languages available in the long run.

What to Expect?

It may look like a straightforward tool, but it brings a whole new experience for music enthusiasts. Google is using Machine Learning to make all this possible. Google explained to its blog how a thing might work out.

“When you hum a melody into Search, our machine learning models transform the audio into a number-based sequence representing the song’s melody. Our models are trained to identify songs based on various 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.

We compare these sequences to thousands of songs from around the world and identify potential matches in real-time. For example, if you listen to Tones and I’s “Dance Monkey,” you’ll recognize the song, whether it was sung, whistled, or hummed. Similarly, our machine learning models recognize the melody of the studio-recorded version of the song, which we can use to match it with a person’s hummed audio.”

 It’s pretty cool and unique, and we all can’t wait to try this out!



You may also like

{"email":"Email address invalid","url":"Website address invalid","required":"Required field missing"}