Publications
Results
- Showing results for:
- Reset all filters
Search results
-
Conference paperLim F, Naylor PA, 2012,
Relaxed multichannel least squares with constrained initial taps for multichannel dereverberation
This paper presents a novel algorithm for robust multichannel dereverberation in the presence of system identification errors with the specific aim of avoiding colouration of the equalized signal. Our proposed algorithm is based upon the technique of channel shortening, which targets only the late taps of the room impulse response. Within the framework of the relaxed multichannel least squares (RMCLS) algorithm, we employ partial relaxation of the early taps of the equalized impulse response (EIR) to increase robustness to channel estimation errors, while constraining the initial taps to avoid undesirable colouration of the equalized signal. It is shown through quantitative experimental results that the resultant equalized signal has an overall improved speech quality perception when compared to alternative algorithms.
-
Conference paperGaubitch ND, Löllmann HW, Jeub M, et al., 2012,
Performance comparison of algorithms for blind reverberation time estimation from speech
The reverberation time, T60, is one of the key parameters used to quantify room acoustics. It can provide information about the quality and intelligibility of speech recorded in a reverberant environment, and it can be used to increase robustness to reverberation of speech processing algorithms. T60 can be determined directly from a measurement of the acoustic impulse response, but in situations where this is unavailable it must be estimated blindly from reverberant speech. In this contribution, we provide a study of three state-of-the-art methods for blind T60 estimation. Experimental results with a large number of talkers, simulated and measured acoustic impulse responses, and various levels of additive white Gaussian noise are presented. The relative merits of the three methods in terms of computational time, estimation accuracy, noise sensitivity and inter-talker variance are discussed. In general, all three methods are able to estimate the reverberation time to within 0.2 s for T60 ≤ 0.8 s and SNR ≥ 30 dB, while increasing the noise level causes overestimation. The relative computational speed of the three methods is also assessed.
-
Conference paperNaylor PA, Gaubitch ND, 2012,
Acoustic signal processing in noise: It's not getting any quieter
Acoustic signal processing research has been addressing the issues associated with additive noise and other degradations in speech for many years and several significant technical advances are now embedded in the state-of-the-art. Nevertheless, the problems are not solved and may actually be worsening. The philosophy advocated in this paper is that further improvements in acoustic signal processing for noise reduction and robustness are, of course, important but are unlikely to be sufficient on their own. Alongside the signal processing, successful systems are likely going to need to include two further factors: an element of matching to the human perception system and also an element of sensing and adaptation to the local environment, giving systems acoustic awareness. Examples of current research on human perception and acoustic signal processing are discussed. These include some aspects of auditory cognition and signal processing methods for building acoustic awareness. A new initiative for benchmarking is also highlighted.
-
Conference paperCanclini A, Antonacci F, Filos J, et al., 2012,
Exact localization of planar acoustic reflectors in three-dimensional geometries
In this paper we propose a methodology for localizing acoustic planar reflectors in a 3D geometry, using acoustic measurements acquired by a set of microphones. An acoustic source emitting a known signal is placed close to the wall to be identified, and is used for estimating the source-to-microphone impulse responses. In a preliminary step, such estimates are employed for localizing the source. After that, the Times Of Arrival (TOAs) associated to the first order reflective paths are extracted from the impulse responses and converted into quadratic constraints (ellipsoids) acting on the reflective plane. The constraints are then collected into acost function, whose exact minimization leads to the searched plane. A theoretical analysis is performed for predicting the impact of measurement errors on the estimation. Moreover, experimental results in a real meeting room prove the effectiveness of the method.
-
Conference paperSharma D, Hilkhuysen G, Naylor PA, et al., 2012,
Descriptive Vocabulary Development for Degraded Speech
, 13th Annual Conference of the International-Speech-Communication-Association, Publisher: ISCA-INT SPEECH COMMUNICATION ASSOC, Pages: 1494-1497 -
Journal articleAntonacci F, Filos J, Thomas M, et al., 2012,
Inference of room geometry from acoustic impulse responses
, IEEE Trans. Audio Speech Language Proc., Vol: 20, Pages: 2683-2695 -
Journal articleAnnibale P, Filos J, Naylor PA, et al., 2012,
TDOA-based speed of sound estimation for air temperature and room geometry inference
, IEEE Trans. Audio, Speech, Lang. Process. -
Journal articleJarrett D, Habets EAP, Thomas M, et al., 2012,
Rigid sphere room impulse response simulation: algorithm and applications
, J. Acoust. Soc. America, Vol: 132 -
Conference paperThomas MRP, Gaubitch ND, Habets EAP, et al., 2012,
AN INSIGHT INTO COMMON FILTERING IN NOISY SIMO BLIND SYSTEM IDENTIFICATION
, IEEE International Conference on Acoustics, Speech and Signal Processing, Publisher: IEEE, Pages: 521-524, ISSN: 1520-6149- Author Web Link
- Cite
- Citations: 2
-
Conference paperSharma D, Naylor PA, Gaubitch ND, et al., 2012,
NON INTRUSIVE CODEC IDENTIFICATION ALGORITHM
, IEEE International Conference on Acoustics, Speech and Signal Processing, Publisher: IEEE, Pages: 4477-4480, ISSN: 1520-6149- Author Web Link
- Cite
- Citations: 7
-
Conference paperFilos J, Canclini A, Antonacci F, et al., 2012,
LOCALIZATION OF PLANAR ACOUSTIC REFLECTORS FROM THE COMBINATION OF LINEAR ESTIMATES
, 20th European Signal Processing Conference (EUSIPCO), Publisher: IEEE COMPUTER SOC, Pages: 1019-1023, ISSN: 2076-1465- Author Web Link
- Cite
- Citations: 14
-
Conference paperAnnibale P, Antonacci F, Bestagini P, et al., 2011,
The SCENIC Project: Space-Time Audio Processing for Environment-Aware Acoustic Sensing and Rendering
-
Journal articleSlaney M, Naylor PA, 2011,
Audio and Acoustic Signal Processing
, IEEE SIGNAL PROCESSING MAGAZINE, Vol: 28, Pages: 160-U26, ISSN: 1053-5888 -
Conference paperLoganathan P, Habets EAP, Naylor PA, 2011,
A Proportionate Adaptive Algorithm with Variable Partitioned Block Length for Acoustic Echo Cancellation
-
Conference paperJarrett DP, Thomas MR, Habets EAP, et al., 2011,
Simulating Room Impulse Responses for Spherical Microphone Arrays
-
Conference paperSharma D, Hilkhuysen G, Gaubitch ND, et al., 2011,
C-Qual - A validation of PESQ using degradations encountered in forensic and law enforcement audio
, Pages: 177-181Assessment of speech quality of law-enforcement audio recordings is important as degradations introduced by non-ideal recording conditions can reduce the intelligence value of such recordings. Furthermore a model that predicts speech quality could be beneficial for assessing the performance of audio collection and enhancement systems. The Perceptual Evaluation of Speech Quality (PESQ) algorithm (ITU-T P.862) has been validated for degradations common in telecommunications. In this paper we apply PESQ to degradations typically encountered in law-enforcement. Also we present a subjectively labeled database (C-Qual) containing distortions encountered in law enforcement scenarios. Comparing the prediction by PESQ and the observed opinions provided by the listeners shows that PESQ is less suitable for estimating the speech quality in this context.
-
Conference paperGaubitch ND, Brookes M, Naylor PA, et al., 2011,
Bayesian Adaptive method for estimating Speech Intelligibility in noise
, Pages: 169-174We present the Bayesian Adaptive Speech Intelligibility Estimation (BASIE) method - a tool for rapid estimation of a given speech reception threshold (SRT) and the slope at that threshold of multiple psychometric functions for speech intelligibility in noise. The core of this tool is an adaptive Bayesian procedure, which adjusts the signal-to-noise ratio at each subsequent stimulus such that the expected variance of the threshold and slope estimates are minimised. Simulation results show that the algorithm is able to achieve SRT estimates accurate to within ±1 dB in under 30 iterations. Furthermore, we discuss strategies for using BASIE to evaluate the effects of speech processing algorithms on intelligibility and we give two illustrative examples for different noise reduction methods with supporting listening experiments.
-
Conference paperGaubitch ND, Brookes M, Naylor PA, et al., 2011,
Single-Microphone Blind Channel Identification in Speech Using Spectrum Classification
-
Journal articleGudnason J, Thomas MRP, Ellis DPW, et al., 2011,
Data-Driven Voice Source Waveform Analysis and Synthesis
, Speech Communication, Vol: to appear -
Conference paperHabets EAP, Benesty J, Naylor PA, 2011,
A Cross-Relation Based Affine Projection Algorithm for Blind SIMO System Identification
This data is extracted from the Web of Science and reproduced under a licence from Thomson Reuters. You may not copy or re-distribute this data in whole or in part without the written consent of the Science business of Thomson Reuters.