BMAT monitors and reports music across televisions, radios, venues, and digital services globally. Guys can tell when and where any song is played and provide the metadata that describes who owns the rights to each track.
We helped BMAT to build a music identification system that actually gives recognition to a dozen artists. If, let’s say, with Shazam you can just identify and upload the music, while this system is processing more than 100 million sales from Youtube, Spotify, Apple Music, Amazon.
Here are some features that were implemented by our engineers:
- Collecting information about the frequency makeup of the signal using NumPy
- Music Recognition using fingerprinting
- Comparing the audio fingerprint of a user-supplied recording
Tech stack we used