Proceedings Article10.1109/ICASSP.2005.1415139
Closely coupled array processing and model-based compensation for microphone array speech recognition
Xianyu Zhao,Zhijian Ou,Minhua Chen,Zuoying Wang +3 more
- 18 Mar 2005
- Vol. 1, pp 417-420
TL;DR: A new microphone array speech recognition system in which the array processor and the speech recognizer are closely coupled is studied, which significantly improved the speech recognition performance in overlapping speech situations.
read more
Abstract: In this paper, a new microphone array speech recognition system in which the array processor and the speech recognizer are closely coupled is studied. The system includes a generalized sidelobe canceller (GSC) beamformer followed by a recognizer with vector Taylor series (VTS) compensation. The GSC beamformer provides two outputs, allowing more information to be used in the recognizer. One is the enhanced target speech output, the other is the reference noise output. VTS is used to compensate the effect of the residual noise in the GSC speech output, utilizing the GSC reference noise output. The compensation is done in a minimum mean square error (MMSE) sense. Moreover, an iteration procedure using an expectation-maximization (EM) algorithm is developed to refine the compensation parameters. Experimental results on the MONC database showed that the new system significantly improved the speech recognition performance in overlapping speech situations.
read more
Chat with Paper
AI Agents for this Paper
Find similar papers on Google Scholar, PubMed and Arxiv
Write a critical review of this paper
Analyze citations of this paper to find unaddressed research gaps
Citations
Patent
Forming beams with nulls directed at noise sources
William V. Oxford
- 13 Apr 2006
TL;DR: In this article, a processor performs a broadside scan on the microphone array and analyzes the resulting amplitude envelope to identify acoustic source angles, which are further investigated with a directed beam (e.g., a hybrid superdirective/delay-and-sum beam) to obtain a corresponding beam signal.
71
Patent
Tracking talkers using virtual broadside scan and directed beams
William V. Oxford
- 11 Apr 2006
TL;DR: In this paper, a processor is configured to perform acoustic echo cancellation, to track multiple talkers with highly directed beams, to design beams with nulls pointed at noise sources, to generate a 3D model of the physical environment, to compensate for the proximity effect, and to perform dereverberation of a talker's voice signal.
26
Applications of Array Signal Processing
TL;DR: The focus of this chapter is on describing many of the various ASP applications that have been studied in the literature, with a particular emphasis on explaining the assumptions made by researchers studying these problems and demonstrating how these assumptions lead to the common data model or something similar.
21
Closely Coupled Array Processing and Model-Based Compensation for Microphone Array Speech Recognition
Xianyu Zhao,Zhijian Ou +1 more
TL;DR: A closely coupled approach is proposed, in which a recognizer with model-based noise compensation exploits the reference noise outputs from a MIMO array processor and significantly improved the speech recognition performance in the overlapping speech situations.
16
MLP-based log spectral energy mapping for robust overlapping speech recognition
Weifeng Li,Mathew Magimai.-Doss,John Dines,Hervé Bourlard +3 more
- 25 Aug 2008
TL;DR: By learning the mapping between log MFBEs extracted from noisy and clean signals the performance of ASR system can be significantly improved in overlapping multi-speaker condition compared a conventional delay-sum beamforming approach, while keeping the performance on single non-overlapping speaker condition intact.
7
References
The generalized correlation method for estimation of time delay
TL;DR: In this paper, a maximum likelihood estimator is developed for determining time delay between signals received at two spatially separated sensors in the presence of uncorrelated noise, where the role of the prefilters is to accentuate the signal passed to the correlator at frequencies for which the signal-to-noise (S/N) ratio is highest and suppress the noise power.
4.8K
An alternative approach to linearly constrained adaptive beamforming
L.J. Griffiths,C. Jim +1 more
TL;DR: A beamforming structure is presented which can be used to implement a wide variety of linearly constrained adaptive array processors and is shown to incorporate algorithms which have been suggested previously for use in adaptive beamforming as well as to include new approaches.
2K
Robust adaptive beamforming
TL;DR: It is shown that a simple scaling of the projection of tentative weights, in the subspace orthogonal to the linear constraints, can be used to satisfy the quadratic inequality constraint.
2K
•Book
Microphone Arrays Signal Processing Techniques and Applications
M.S. Brandstein,Darren Ward +1 more
- 01 Jan 2001
TL;DR: This paper presents a meta-modelling architecture for microphone Array Processing that automates the very labor-intensive and therefore time-heavy and expensive process of manually shaping Microphone Arrays for Speech Input in Automobiles.
1.4K