The World of Sound Effects

Reading Time: 5 minutesWhen we started to develop our audio fingerprinting technology, we made sure it worked with every sound we recorded but we still had a way ahead to delve into its nuts and bolts and overcome some of the oddities of our industry. This leads us on to the fascinating world of sound effects. Some weeks…

Keep Learning Deep Learning

Reading Time: 4 minutesYou never know when you’ll close the loop. After a brief encounter with Neural Networks, Àlex decided to go find an algorithm somewhere else. Two decades later they met again and became pals. Now called Deep Learning, Neural Networks showed up last year at BMAT to help us solve one of our biggest challenges: teach…

How We Identify Songs. Audio Fingerprinting

Reading Time: 4 minutesOur motto is “We hear everything everywhere and tell everyone who wants to know.” Something we don’t mention so often but we feel equally proud of is how we do it: audio fingerprinting. It’s literally our identity sign, the system that allows us to identify 52 years of audio against 60M sound recordings everyday. Tempo…

BMAT’s Melody Identification – Part 2: Running Out of Melodies

Reading Time: 3 minutesSee Part 1 – Music and Language. The long awaited second part is here. No matter how you represent a melody, as long as the representation uses limited and quantised metrics – like a music score over a certain number of bars does – the number of possible melodies is finite. Despite infinite expression nuances…

BMAT’s Melody Identification – Part 1: Music and Language

Reading Time: 3 minutesSee Part 2 – Running Out of Melodies Years ago we investigated whether language and music were the same. In the process we found out that the universe of tunes behave as a social network and that western popular music might be running out of unique melodies. Men sung before they talked. Apparently our innate…