Extending Deep Rhythm for Tempo and Genre Estimation Using Complex Convolutions, Multitask Learning and Multi-input Network

Hadrien Foroughmand (1) , Geoffroy Peeters (2, 3, 4, 5)

1 Analyse et synthèse sonores [Paris]
2 LTCI - Laboratoire Traitement et Communication de l'Information
3 IDS - Département Images, Données, Signal
4 S2A - Signal, Statistique et Apprentissage
5 IP Paris - Institut Polytechnique de Paris

Tempo and genre are two inter-leaved aspects of music, genres are often associated to rhythm patterns which are played in specific tempo ranges. In this paper, we focus on the recent Deep Rhythm system based on a harmonic representation of rhythm used as an input to a convolutional neural network. To consider the relationships between frequency bands, we process complex-valued inputs through complexconvolutions. We also study the joint estimation of tempo/genre using a multitask learning approach. Finally, we study the addition of a second input branch to the system based on a VGG-like architecture applied to a mel-spectrogram input. This multi-input approach allows to improve the performances for tempo and genre estimation.

Mots clés: Tempo estimation; genre classification; deep-learning; complex network; multitask; multi-input

Domaines: Son [cs.SD]; Intelligence artificielle [cs.AI]; Traitement du signal et de l'image [eess.SP]; Musique, musicologie et arts de la scène

Hadrien Foroughmand, Geoffroy Peeters. Extending Deep Rhythm for Tempo and Genre Estimation Using Complex Convolutions, Multitask Learning and Multi-input Network. The 2020 Joint Conference on AI Music Creativity, Bob Sturm, Oct 2020, Stockholm, Sweden. ⟨hal-03127155⟩