Skip to Main content Skip to Navigation

Solving global permutation ambiguity of time domain BSS using speaker specific features of speech signals

Abstract : Multidimensional localization of multiple sources using BSS based TDOA estimators, requires the solution of global permutation ambiguity before fusing several TDOA estimations. Since the separation quality of BSS isn't always perfect, it is not easy to decide which TDOA belongs to which source. Here we study the possibility of using several speaker specific features of speech signal in order to recognize perceptually dominant sources in each one of moderately separated outputs of BSS algorithm. We compare the feasibility of different features in terms of validity rate of decisions and computational complexity.
Document type :
Conference papers
Complete list of metadatas

Cited literature [12 references]  Display  Hide  Download

https://hal.inria.fr/hal-00938356
Contributor : H. Yahia <>
Submitted on : Wednesday, January 29, 2014 - 12:43:51 PM
Last modification on : Wednesday, November 29, 2017 - 9:22:57 AM
Document(s) archivé(s) le : Monday, May 5, 2014 - 11:06:35 AM

File

vk2.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-00938356, version 1

Citation

Vahid Khanagha, Ali Khanagha. Solving global permutation ambiguity of time domain BSS using speaker specific features of speech signals. 2009 IEEE Symposium on Industrial Electronics and Applications (ISIEA 2009), IEEE, Oct 2009, Kuala Lumpur, Malaysia. ⟨hal-00938356⟩

Share

Metrics

Record views

141

Files downloads

278