Feature Adapted Convolutional Neural Networks for Downbeat Tracking

Abstract :

We define a novel system for the automatic estimation of downbeat positions from audio music signals. New rhythm and melodic features are introduced and feature adapted convolutional neural networks are used to take advantage of their specificity. Indeed, invariance to melody transposition, chroma data augmentation and length-specific rhythmic patterns prove to be useful to learn downbeat likelihood. After the data is segmented in tatums, complementary features related to melody, rhythm and harmony are extracted and the likelihood of a tatum being at a downbeat position is computed with the aforementioned neural networks. The downbeat sequence is then extracted with a flexible temporal hidden Markov model. We then show the efficiency and robustness of our approach with a comparative evaluation conducted on 9 datasets.

Complete list of metadatas

https://hal.telecom-paristech.fr/hal-02287268
Contributor : Telecomparis Hal <>
Submitted on : Friday, September 13, 2019 - 4:47:17 PM
Last modification on : Thursday, October 17, 2019 - 12:37:02 PM

Identifiers

  • HAL Id : hal-02287268, version 1

Citation

Simon Durand, Juan P. Bello, Bertrand David, Gael Richard. Feature Adapted Convolutional Neural Networks for Downbeat Tracking. ICASSP 2016, Sep 2016, Shanghai, China. ⟨hal-02287268⟩

Share

Metrics

Record views

3