Multi-Pitch Estimation using NHF with Multi-Dictionary Distinguishing Attack and Reverberation of Sounds
- 1 November 2019
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 401, 1836-1841
Abstract
This paper proposes a multiple pitch estimation algorithm for the piano music which improves both precision and processing time. The conventional method applies non-negative matrix factorization (NMF) and singular value decomposition to ensure the continuity of sound and pattern of musics. However, audio signal spectrogram has large elements and processing singular value decomposition is inefficient in the computational time. Our method separates the input audio spectrogram into the block of sound. Then we apply a NMF with group sparsity constraint to enforce the continuity of sound. In addition, we use the different dictionaries for the attack part and the reverberation part of the sound. It improves the precision of the estimation for each part.Keywords
This publication has 9 references indexed in Scilit:
- Automatic music transcription using low rank non-negative matrix decomposition2017
- NMF-based multiple pitch estimation using sparseness and inter-frame continuity constraints2014
- Multipitch Estimation of Piano Sounds Using a New Probabilistic Spectral Smoothness PrincipleIEEE Transactions on Audio, Speech, and Language Processing, 2009
- Adaptive Harmonic Spectral Decomposition for Multiple Pitch EstimationIEEE Transactions on Audio, Speech, and Language Processing, 2009
- Non-local sparse models for image restoration2009
- Nonnegative Matrix Factorization with the Itakura-Saito Divergence: With Application to Music AnalysisNeural Computation, 2009
- Automatic Transcription of Melody, Bass Line, and Chords in Polyphonic MusicComputer Music Journal, 2008
- A Generalized Divergence Measure for Nonnegative Matrix FactorizationNeural Computation, 2007
- Learning the parts of objects by non-negative matrix factorizationNature, 1999