search

worldveil / dejavu

Audio fingerprinting and recognition in Python
MIT License
6.41k stars 1.43k forks source link

Added audio length (in seconds) to database. Added relative confidence value #148

Closed JPery closed 5 years ago

JPery commented 6 years ago

Now we can know the length of the song that we have recognized

JPery commented 6 years ago

This also resolves the issue #132 created by me

pk97 commented 6 years ago

Thanks Jpery!

JPery commented 6 years ago

I've added a "relative confidence" value based on the number of fingerprints to match and the matched fingerprints. After studying this value with some audios, we have achieve a value that only increases if the match is correct. We can see an example in this image:

image

In the graph, the blue line represents an audio NOT in the database and the green one represents a part of an audio in the database.

JPery commented 5 years ago

This request is continued in #197