Acoustic Source Separation of Convolutive Mixtures Based on Intensity Vector Statistics

Günel, Banu; Hachabiboglu, H.; Kondoz, A.M.

doi:10.1109/tasl.2008.918967

Cited by 51 publications

(51 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…BSS techniques using single spatial microphone recordings have been proposed in [3], [9] where time-frequencies whose DOA estimates (correspond to peaks in a histogram of DOA) formed for a speech segment are grouped or clustered together to form sources. However, the results of Fig.…”

Section: B Problems Of Single Spatial Microphone Bssmentioning

confidence: 99%

“…Speech Source DOA Estimation Similar to [3], [9], the microphone array used in this work records in B-format and consists of one omnidirectional (W) and three figure-of-eight directional (X, Y, Z) channels. These channels are firstly transferred to the time-frequency domain using an MDCT [10].…”

Section: Collaborative Blind Source Separationmentioning

confidence: 99%

“…Thus, the DOA of each speech source can be obtained by examining the DOA histogram of these time-frequency instants [3], [9]. Here the aim is to examine the BSS algorithm assuming perfect knowledge of the source DOAs.…”

Section: Collaborative Blind Source Separationmentioning

confidence: 99%

“…Extended techniques have been proposed to separate convolutive mixtures [2]. These stochastic-based methods generally suffer from computational expenses [3] especially for highly convolutive mixtures. Sparse-based approaches assume approximate W-disjoint orthogonality [4] of the speech signals and separate the simultaneous speech sources by grouping together time-frequency components belonging to the same speech source and are more computationally efficient compared to stochastic-based methods [3].…”

Section: Introductionmentioning

confidence: 99%

“…Sparse-based approaches assume approximate W-disjoint orthogonality [4] of the speech signals and separate the simultaneous speech sources by grouping together time-frequency components belonging to the same speech source and are more computationally efficient compared to stochastic-based methods [3]. Such grouping can be based on time and phase delays [4] obtained from processing spaced microphone array recordings or intensity-based Direction of Arrival (DOA) estimates obtained from co-located (spatial) microphone recordings and using microphone directivities [3].…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Collaborative Blind Source Separation Using Location Informed Spatial Microphones

Zheng

Ritz

2013

IEEE Signal Process. Lett.

View full text Add to dashboard Cite

, "Collaborative blind source separation using location informed spatial microphones," IEEE Signal Processing Letters, vol. 20, (1) pp. 83-86, 2013. Collaborative blind source separation using location informed spatial microphones AbstractThis letter presents a new Collaborative Blind Source Separation (CBSS) technique that uses a pair of location informed coincident microphone arrays to jointly separate simultaneous speech sources based on timefrequency source localization estimates from each microphone recording. While existing BSS approaches are based on localization estimates of sparse time-frequency components, the proposed approach can also recover non-sparse (overlapping) time-frequency components. The proposed method has been evaluated using up to three simultaneous speech sources under both anechoic and reverberant conditions. Results from objective and subjective measures of the perceptual quality of the separated speech show that the proposed approach significantly outperforms existing BSS approaches. © 1994-2012 IEEE. The proposed method has been evaluated using up to three simultaneous speech sources under both anechoic and reverberant conditions. Results from objective and subjective measures of the perceptual quality of the separated speech show that the proposed approach significantly outperforms existing BSS approaches.

show abstract

Section: B Problems Of Single Spatial Microphone Bssmentioning

confidence: 99%