Speaker Recognition In Noisy Conditions With Limited Training Data

Research output: Contribution to conferencePaper

2 Citations (Scopus)
60 Downloads (Pure)

Abstract

In this paper we present a novel method for performing speaker recognition with very limited training data and in the presence of background noise. Similarity-based speaker recognition is considered so that speaker models can be created with limited training speech data. The proposed similarity is a form of cosine similarity used as a distance measure between speech feature vectors. Each speech frame is modelled using subband features, and into this framework, multicondition training and optimal feature selection are introduced, making the system capable of performing speaker recognition in the presence of realistic, time-varying noise, which is unknown during training. Speaker identi?cation experiments were carried out using the SPIDRE database. The performance of the proposed new system for noise compensation is compared to that of an oracle model; the speaker identi?cation accuracy for clean speech by the new system trained with limited training data is compared to that of a GMM trained with several minutes of speech. Both comparisons have demonstrated the effectiveness of the new model. Finally, experiments were carried out to test the new model for speaker identi?cation given limited training data and with differing levels and types of realistic background noise. The results have demonstrated the robustness of the new system.
Original languageEnglish
Pages1294-1298
Number of pages5
Publication statusPublished - Sep 2011
Event19th European Signal Processing Conference - Barcelona, Spain
Duration: 01 Sep 201101 Sep 2011
http://www.eurasip.org/Proceedings/Eusipco/Eusipco2011/index.html

Conference

Conference19th European Signal Processing Conference
Abbreviated titleEUSIPCO 2011
CountrySpain
CityBarcelona
Period01/09/201101/09/2011
OtherThe 2011 European Signal Processing Conference (EUSIPCO-2011) has been the nineteenth in a series of conferences promoted by the European Association for Signal Processing (EURASIP). This year edition has taken place in Barcelona, capital city of Catalonia (Spain), from August 29 to September 2, and has been jointly organized by the Centre Tecnològic de Telecomunicacions de Catalunya (CTTC) and the Universitat Politècnica de Catalunya (UPC).
Internet address

Fingerprint Dive into the research topics of 'Speaker Recognition In Noisy Conditions With Limited Training Data'. Together they form a unique fingerprint.

Cite this