Training Ircam's Score Follower

Arshia Cont (1) , Diemo Schwarz (1) , Norbert Schnell (1)

1 Equipe Interactions musicales temps-réel

This paper describes our attempt to make the Hidden Markov Model (HMM) score following system developed at Ircam sensible to past experiences in order to obtain better audio to score real-time alignment for musical applications. A new observation modeling based on Gaussian Mixture Models is developed which is trainable using a learning algorithm we would call automatic discriminative training. The novelty of this system lies in the fact that this method, unlike classical methods for HMM training, is not concerned with modeling the music signal but with correctly choosing the sequence of music events that was performed. Besides obtaining better alignment, new system's parameters are controllable in a physical manner and the training algorithm learns different styles of music performance as discussed.

Mots clés: score following; Hidden Markov Models; automatic accompaniment; performing; training; learning; Gaussian Mixture Models; probabilistic modeling

Mots clés: Informatique musicale

Domaines: Son [cs.SD]; Interface homme-machine [cs.HC]; Musique, musicologie et arts de la scène; Traitement du signal et de l'image [eess.SP]; Apprentissage [cs.LG]; Intelligence artificielle [cs.AI]; Ingénierie assistée par ordinateur; Multimédia [cs.MM]; Vision par ordinateur et reconnaissance de formes [cs.CV]; Autre [cs.OH]; Traitement du signal et de l'image [eess.SP]

Arshia Cont, Diemo Schwarz, Norbert Schnell. Training Ircam's Score Follower. IEEE International Conference on Acoustics, Speech, and Signal Processing, Mar 2005, Philadelphia, United States. pp.1-1. ⟨hal-01161340⟩