Home | Sitemap | deutsch  | Legals | Data Protection | KIT

How does music recognition actually work and what does it have to do with math?

https://wsm6.scc.kit.edu/CMS/ioLoadImage.asp?Action=ShowPreview&SubFolderGuid=3ABC14B8057440CA8CE0D10C0D5D0051&ImageTitle=Shazam.png&FolderGuid=3ABC14B8057440CA8CE0D10C0D5D0051&DummyRefreshTime=1515403151838
Source: tinyurl.com/y8cpkk2s

Shazam is an app for the smartphone that detects unknown songs in fractions of a second and tells the user everything about the song title and songwriter. Shazam is now one of the most popular and popular apps in the world - the database is used up to 4 million times a day! Especially the speed with which songs are found and the enormously large database of available music are what make Shazam so special. But what does this have to do with math?

What makes Shazam so successful is the idea to generate an acoustic fingerprint of a song. This fingerprint is, just like human beings, unique for each piece of music. The high efficiency of the acoustic fingerprint algorithm as well as an intelligent procedure for searching the database have made Shazam successful.

The exact mathematical steps behind the fingerprint algorithm and the search of the Shazam database will be explained in the Shazam-CAMMP day very closely and interactively. Using simple MATLAB codes, students discover the mathematical basics of sound and the mathematical tool for acoustic fingerprinting - Fourier analysis. In an action-oriented module, the students will create an acoustic fingerprint of a sample photograph themselves and search it in a programmed database. This only requires prior knowledge of functions.

Duration: from 5 hours (incl. lunch break)
Contents: Functional equations, trigonometric functions
Previous knowledge: functional concept
Participants: Mathematics courses from grade 9 upwards
Created by: Jonas Kusch, Nils Steffens, Janna Tinnes
Registration: Appointments can be arranged individually by e-mail.