Blind Source Separation for Moving Speech Signals Using Blockwise ICA and Residual Crosstalk Subtraction(Speech/Acoustic Signal Processing)(<Special Section>Digital Signal Processing)
スポンサーリンク
概要
- 論文の詳細を見る
This paper describes a real-time blind source separation (BSS) method for moving speech signals in a room. Our method employs frequency domain independent component analysis (ICA) using a blockwise batch algorithm in the first stage, and the separated signals are refined by postprocessing using crosstalk component estimation and non-stationary spectral subtraction in the second stage. The blockwise batch algorithm achieves better performance than an online algorithm when sources are fixed, and the postprocessing compensates for performance degradation caused by source movement. Experimental results using speech signals recorded in a real room show that the proposed method realizes robust real-time separation for moving sources. Our method is implemented on a standard PC and works in realtime.
- 社団法人電子情報通信学会の論文
- 2004-08-01
著者
-
Sawada Hiroshi
Ntt Communication Science Laboratories Ntt Corporation
-
Makino Shoji
NTT Communication Science Laboratories
-
MUKAI Ryo
NTT Communication Science Laboratories
-
Makino Shoji
Ntt Communication Science Laboratories Ntt Corporation
-
Makino S
Ntt Communication Sci. Lab. Kyoto Jpn
-
Mukai Ryo
Ntt Communication Science Laboratories Ntt Corporation
-
ARAKI Shoko
NTT Communication Science Laboratories, NTT Corporation
-
Araki Shoko
Ntt Communication Science Laboratories Ntt Corporation
-
Sawada Hiroshi
Ntt Communication Science Laboratories
関連論文
- A Fast Projection Algorithm for Adaptive Filtering
- A desing of a hands-free communication unit using loudspeakers and microphones with a flat directional pattern
- Multistage SIMO-Model-Based Blind Source Separation Combining Frequency-Domain ICA and Time-Domain ICA(Adaptive Signal Processing and Its Applications)
- A General Framework to Use Various Decomposition Methods for LUT Network Synthesis
- Efficient Kernel Generation Based on Implicit Cube Set Representations and Its Applications (Special Section on VLSI Design and CAD Algorithms)
- Restructuring Logic Representations with Simple Disjunctive Decompositions (Special Section on VLSI Design and CAD Algorithms)
- Subjective Assessment of the Desired Echo Return Loss for Subband Acoustic Echo Cancellers
- An Efficient Method for Finding an Optimal Bi-Decomposition (Special Section on VLSI Design and CAD Algorithms)
- Logic Synthesis for Look-Up Table Based FPGAs Using Functional Decomposition and Boolean Resubstitution (Special Issue on Synthesis and Verification of Hardware Design)
- FOREWORD
- Subband-Based Blind Separation for Convolutive Mixtures of Speech(Engineering Acoustics)
- Estimating the number of sources using independent component analysis
- Blind Source Separation of Convolutive Mixtures of Speech in Frequency Domain(Multi-channel Acoustic Signal Processing)
- Blind Source Separation for Moving Speech Signals Using Blockwise ICA and Residual Crosstalk Subtraction(Speech/Acoustic Signal Processing)(Digital Signal Processing)
- Convolutive blind source separation for more than two sources in the frequency domain
- Evaluation of separation and dereverberation performance in frequency domain blind source separation
- Underdetermined Blind Separation of Convolutive Mixtures of Speech Using Time-Frequency Mask and Mixing Matrix Estimation(Blind Source Separation, Multi-channel Acoustic Signal Processing)
- Sparse source separation based on simultaneous clustering of source locational and spectral features
- Stereophonic acoustic echo cancellation : An overview and recent solutions
- Polar Coordinate Based Nonlinear Function for Frequency-Domain Blind Source Separation