Audio Source Separation by Probabilistic Latent Component Analysis
نویسندگان
چکیده
The problem of audio source separation from a monophonic sound mixture having known instrument types but unknown timbres is presented. An improvement to the Probabilistic Latent Component Analysis (PLCA) source separation method is proposed. The technique uses a basis function dictionary to produce a first round PLCA source separation. The PLCA weights are then refined by incorporating note onset information. The source separation is then performed using a second round PLCA in which the refined weights are held fixed, and the basis functions are updated. Preliminary experimental results on mixtures of two instruments are quite promising, showing a 6 dB improvement in SIR over standard PLCA.
منابع مشابه
Elementary Sources: Latent Component Analysis for Music Composition
Complexity of music audio signals creates an access problem to specific musical objects or structures within the source samples. Instead of employing more commonly used audio analysis or production techniques to access features, we describe extraction of sub-mixtures from real-world audio using a Probabilistic Latent Component Analysis-based decomposition tool for music composition. This is hig...
متن کاملA Computational Model for Multi - Instrument Music Transcription CS 229 Final Project Report , Autumn 2013
The aim of our project is to build a model for multi-instrument music transcription. Automatic music transcription is the process of converting an audio wave file into some form of music notes representations. We propose a two-step process for an automatic multiinstrument music transcription system including timbre classification and source separation using probabilistic latent component analysis.
متن کاملProbabilistic Latent Component Analysis for Gearbox Vibration Source Separation
Probabilistic latent component analysis (PLCA) is applied to the problem of gearbox vibration source separation. A model for the probability distribution of gearbox vibration employs a latent variable intended to correspond to a particular vibration source, with the measured vibration at a particular sensor for each source the product of a marginal distribution of vibration by frequency, a marg...
متن کاملScale-invariant probabilistic latent component analysis Analyse probabiliste en composantes latentes invariante par homothétie
2011D003 1 Scale-invariant probabilistic latent component analysis Analyse probabiliste en composantes latentes invariante par homothétie Résumé—Dans cette article, nous présentons une nouvelle méthode de décomposition de spectrogrammes musicaux. Cette méthode vise à transposer les decom-positions invariantes par translation qui permettent de décomposer des spectrogrammes à Q-constant (avec une...
متن کاملBlind Audiovisual Source Separation Using Sparse Redundant Representations
In this work, we present a method that jointly separates active audio and visual structures on a given mixture. This new concept, the Blind Audiovisual Source Separation (BAVSS), is achieved by exploiting the coherence existing between the recorded signal of a video camera and only one microphone. An efficient representation of audio and video sequences allows to build robust audiovisual relati...
متن کامل