Answer the question
In order to leave comments, you need to log in
Answer the question
In order to leave comments, you need to log in
Digital Signal Processing is a tricky subject, apart from the difficulties of music recognition.
Read about how recognition is done in the article
In short, the essence is reduced to obtaining a digital "fingerprint" of a musical composition on a short sample (up to 15 seconds). This fingerprint is then looked up in a hash table in a set of pre-known fingerprints of different compositions. And now let's imagine: how many musical compositions exist and how to take the right imprint from all of them? Even if you have a huge collection of studio quality and take prints from all the songs, then from the side of the user, say, a mobile phone, there will be a variety of noises: conversations, traffic noise, and others. For DSP, the FFTW
library (FFT) is often used - just one of many types of digital signal analysis.
Didn't find what you were looking for?
Ask your questionAsk a Question
731 491 924 answers to any question