How to identify a song by humming on Android and iPhone
Translated from Spanish and summarized by DistantNews. Read the original for the full story.
At a glance
- Google’s song-search feature lets users hum or whistle for 10 to 15 seconds to find a matching track on Android and iPhone.
- The system uses machine learning to convert the audio into a numerical representation of the melody and compare it with recordings in its database.
- Shazam and SoundHound offer alternative tools for identifying music through ambient audio or voice.
Forgotten song lyrics are no longer essential for finding a track. Google’s app can identify a melody after users hum or whistle it for 10 to 15 seconds on an Android phone or iPhone.
To use the feature, open the Google app, tap the microphone icon, and select “Search a song.” The system then processes the recording and offers possible matches within seconds. Users do not need perfect pitch or the exact lyrics, because the algorithm analyzes the sequence of notes.
Google’s “Hum to Search” feature uses machine learning to turn the captured audio into a numerical sequence representing the melody. The process removes instruments and vocal nuances, allowing the system to compare the underlying musical structure with thousands of recordings from around the world.
Virtual assistants can also respond to commands such as “What song is this?” Other mobile tools, including Shazam and SoundHound, identify songs through a touch button using ambient audio or voice. These services can recognize music playing through speakers or inside other applications, including when headphones are connected.
Originally published by El Comercio in Spanish. Translated, summarized, and contextualized automatically by DistantNews, with a note on how the source frames the story. Not individually reviewed before publishing. How this works.