Search

Abstract: We present the outcomes of three recent evaluation campaigns in the field of audio and biomedical source separation. These campaigns have witnessed a boom in the range of applications of source separation systems in the last few years, as shown by the increasing number of datasets from 1 to 9 and the increasing number of submissions from 15 to 34. We first discuss their impact on the definition of a reference evaluation methodology, together with shared datasets and software. We then present the key results obtained over almost all datasets. We conclude by proposing directions for future research and evaluation, based in particular on the ideas raised during the related panel discussion at the Ninth International Conference on Latent Variable Analysis and Signal Separation (LVA/ICA 2010). [Copyright &y& Elsevier]

This paper proposes a method for estimating the direction of arrival (DOA) of multiple source signals for an underdetermined situation, where the number of sources N exceeds the number of sensors M ( M < N). Some DOA estimation methods have already been proposed for underdetermined cases. However, since most of them restrict their microphone array arrangements, their DOA estimation ability is limited to a 2-dimensional plane. To deal with an underdetermined case where sources are distributed arbitrarily, we propose a method that can employ a 2- or 3-dimensional sensor array. Our new method employs the source sparseness assumption to handle an underdetermined case. Our formulation with the sensor coordinate vectors allows us to employ arbitrarily arranged sensors easily. We obtained promising experimental results for 2-dimensionally distributed sensors and sources 3×4, 3×5 (#sensors × #speech sources), and for 3-dimensional case with 4×5 in a room (reverberation time (RT) of 120 ms). We also investigate the DOA estimation performance under several reverberant conditions. [ABSTRACT FROM AUTHOR]

This paper proposes a microphone array-based speech activity detection (SAD) method for analyzing multi-party conversations recorded in the presence of noise. In particular, the proposed method considers conversations where the number of speakers and speaker locations cannot be restricted, such as when standing and talking, and at poster sessions. When we observe such conversations, there are directional noise sources and diffuse noise that affect the direction of arrival estimations of the target speech signals. To detect speech activity without a priori knowledge about the speakers and noise environments, a likelihood ratio test (LRT)-based SAD method is applied to spatial magnitude, which are estimated by using the time–frequency masking of the observed spectra. The proposed method can exploit the enhanced signals obtained from time–frequency masking, and works even in the presence of environmental noise. Experiments with recorded simulated poster sessions confirmed that the proposed method could outperform conventional methods based on the LRT for a single channel, magnitude coherence, or crosspower spectrum phase. [ABSTRACT FROM PUBLISHER]

Abstract: This paper presents a new method for blind sparse source separation. Some sparse source separation methods, which rely on source sparseness and an anechoic mixing model, have already been proposed. These methods utilize level ratios and phase differences between sensor observations as their features, and they separate signals by classifying them. However, some of the features cannot form clusters with a well-known clustering algorithm, e.g., the k-means. Moreover, most previous methods utilize a linear sensor array (or only two sensors), and therefore they cannot separate symmetrically positioned sources. To overcome such problems, we propose a new feature that can be clustered by the k-means algorithm and that can be easily applied to more than three sensors arranged non-linearly. We have obtained promising results for two- and three-dimensionally distributed speech separation with non-linear/non-uniform sensor arrays in a real room even in underdetermined situations. We also investigate the way in which the performance of such methods is affected by room reverberation, which may cause the sparseness and anechoic assumptions to collapse. [Copyright &y& Elsevier]

This paper presents a method for enhancing target sources of interest and suppressing other interference sources. The target sources are assumed to be close to, sensors, to have dominant powers at these sensors, and to have non-Gaussianity. The enhancement is performed blindly, i.e., without knowing the position and active time of each source. We consider a general case where the total number of sources is larger than the number of sensors, and neither the number of target sources nor the total number of sources is known. The method is based on a two-stage process where independent component analysis (ICA) is first employed in each frequency bin and then time-frequency masking is used to improve the performance further. We propose a new sophisticated method for deciding the number of target sources and then selecting their frequency components. We also propose a new criterion for specifying time-frequency masks. Experimental results for simulated cocktail party situations in a room, whose reverberation time was 130 ms, are presented to show the effectiveness and characteristics of the proposed method. [ABSTRACT FROM AUTHOR]

Blind source separation (BSS) for convolutive mixtures can be solved efficiently in the frequency domain, where independent component analysis (ICA) is performed separately in each frequency bin. However, frequency-domain BSS involves a permutation problem: the permutation ambiguity of ICA in each frequency bin should be aligned so that a separated signal in the time-domain contains frequency components of the same source signal. This paper Presents a robust and precise method for solving the permutation problem. It is based on two approaches: direction of arrival (DOA) estimation for sources and the interfrequency correlation of signal envelopes. We discuss the advantages and disadvantages of the two approaches, and integrate them to exploit their respective advantages. Furthermore, by utilizing the harmonics of signals, we make the new method robust even for low frequencies where DOA estimation is inaccurate. We also present a new closed-form formula for estimating DOAs from a separation matrix obtained by ICA. Experimental results show that our method provided an almost perfect solution to the permutation problem for a case where two sources were mixed in a room whose reverberation time was 300 ms. [ABSTRACT FROM AUTHOR]

Frequency-domain blind source separation (BSS) is shown to be equivalent to two sets of frequency-domain adaptive beamformers (ABFs) under certain conditions. The zero search of the off-diagonal components in the BSS update equation can be viewed as the minimization of the mean square error in the ABFs. The unmixing matrix of the BSS and the filter coefficients of the ABFs converge to the same solution if the two source signals are ideally independent. If they are dependent, this results in a bias for the correct unmixing filter coefficients. Therefore, the performance of the BSS is limited to that of the ABF if the ABF can use exact geometric information. This understanding gives an interpretation of BSS from a physical point of view. [ABSTRACT FROM AUTHOR]

Two types of inquiry were used to explore the norms governing the management of compliments in Japan and the United States. The first, a semistructured interview, revealed great differences in the frequency of complimenting behavior and in the hierarchy of attributes praised. Incidents from these interviews were used to construct a Complimentary Mode Questionnaire that was then administered to larger samples of Japanese and Americans. The resulting data indicated there were highly significant differences in the reported praising behavior of members of the two cultures. These differences appeared with regard to the status of communicative partners and with regard to the attributes admired. No sex differences were found. When the Japanese were expected to prefer more indirect forms of praise they also showed greater preference for many direct forms; when Americans were expected to prefer more direct forms of praise, they showed surprising preference for indirect forms. The relations of these patterns to other cultural variables and their implications for interaction between the two cultures are suggested. [ABSTRACT FROM PUBLISHER]

• A new objective measure for speech intelligibility is proposed. • The proposed model is based on a signal-to-distortion ratio in the auditory envelope. • Evaluation is performed with speech signals enhanced by nonlinear processing. • The proposed model can predict human results more accurate than conventional models. In this study, we propose a new concept, the gammachirp envelope distortion index (GEDI), based on the signal-to-distortion ratio in the auditory envelope, SDR env , to predict the intelligibility of speech enhanced by nonlinear algorithms. The objective of GEDI is to calculate the distortion between enhanced and clean-speech representations in the domain of a temporal envelope extracted by the gammachirp auditory filterbank and modulation filterbank. We also extend GEDI with multi-resolution analysis (mr-GEDI) to predict the speech intelligibility of sounds under non-stationary noise conditions. We evaluate GEDI in terms of the speech intelligibility predictions of speech sounds enhanced by a classic spectral subtraction and a Wiener filtering method. The predictions are compared with human results for various signal-to-noise ratio conditions with additive pink and babble noises. The results showed that mr-GEDI predicted the intelligibility curves better than short-time objective intelligibility (STOI) measure, extended-STOI (ESTOI) measure, and hearing-aid speech perception index (HASPI) under pink-noise conditions, and better than HASPI under babble-noise conditions. The mr-GEDI method does not present an overestimation tendency and is considered a more conservative approach than STOI and ESTOI. Therefore, the evaluation with mr-GEDI may provide additional information in the development of speech enhancement algorithms. [ABSTRACT FROM AUTHOR]

Abstract: Research on noise robust speech recognition has mainly focused on dealing with relatively stationary noise that may differ from the noise conditions in most living environments. In this paper, we introduce a recognition system that can recognize speech in the presence of multiple rapidly time-varying noise sources as found in a typical family living room. To deal with such severe noise conditions, our recognition system exploits all available information about speech and noise; that is spatial (directional), spectral and temporal information. This is realized with a model-based speech enhancement pre-processor, which consists of two complementary elements, a multi-channel speech–noise separation method that exploits spatial and spectral information, followed by a single channel enhancement algorithm that uses the long-term temporal characteristics of speech obtained from clean speech examples. Moreover, to compensate for any mismatch that may remain between the enhanced speech and the acoustic model, our system employs an adaptation technique that combines conventional maximum likelihood linear regression with the dynamic adaptive compensation of the variance of the Gaussians of the acoustic model. Our proposed system approaches human performance levels by greatly improving the audible quality of speech and substantially improving the keyword recognition accuracy. [Copyright &y& Elsevier]