Efficient iterative mean shift based cosine dissimilarity for multi-recording speaker clustering

Author

Senoussaoui, Mohammed ; Kenny, P. ; Dumouchel, P. ; Stafylakis, Themos

Author_Institution

Ecole de Technol. Super. (ETS), Montréal, QC, Canada

fYear

2013

Firstpage

7712

Lastpage

7715

Abstract

Speaker clustering is an important task in many applications such as Speaker Diarization as well as Speech Recognition. Speaker clustering can be done within a single multi-speaker recording (Diarization) or for a set of different recordings. In this work we are interested by the former case and we propose a simple iterative Mean Shift (MS) algorithm to deal with this problem. Traditionally, MS algorithm is based on Euclidean distance. We propose to use the Cosine distance in order to build a new version of MS algorithm. We report results as measured by speaker and cluster impurities on NIST SRE 2008 datasets.

Keywords

audio recording; iterative methods; speaker recognition; Euclidean distance; MS algorithm; NIST SRE 2008 datasets; cluster impurities; cosine distance; iterative mean shift based cosine dissimilarity; multirecording speaker clustering; single multispeaker recording; speaker diarization; speaker impurities; speech recognition; Clustering algorithms; Euclidean distance; Impurities; Integrated circuits; Kernel; Speaker recognition; Speech; Cluster Impurity; Cosine distance; Mean Shift (MS); Speaker Clustering; Speaker Impurity;

fLanguage

English

Publisher

ieee

Conference_Titel

Acoustics, Speech and Signal Processing (ICASSP), 2013 IEEE International Conference on

Conference_Location

Vancouver, BC

ISSN

1520-6149

Type

conf

DOI

10.1109/ICASSP.2013.6639164

Filename

6639164