Speaker Characterization with MLSFs (seminar) - Revision history

David at 17:34, 12 June 2006

2006-06-12T17:34:26Z

← Older revision		Revision as of 17:34, 12 June 2006
Line 14:		Line 14:
	When combined with mel LSFs differences, MLSFs feature overcomes the performance of the MFCCs for male and female speakers, even with temporal delta, ΔMFCCs, included. Performance is measured in the context of speaker verification, using EER and minimum HTER. Detection error threshold (DET) curves are also presented, as well as HTER curves.		When combined with mel LSFs differences, MLSFs feature overcomes the performance of the MFCCs for male and female speakers, even with temporal delta, ΔMFCCs, included. Performance is measured in the context of speaker verification, using EER and minimum HTER. Detection error threshold (DET) curves are also presented, as well as HTER curves.
	The main objective of this study is to compare different features performances with a common framework, from what a standard support vector machine recogniser was developed.		The main objective of this study is to compare different features performances with a common framework, from what a standard support vector machine recogniser was developed.

	Tests are based on the cellular component of the “2002 NIST Speaker Recognition Evaluation Corpus		Tests are based on the cellular component of the “2002 NIST Speaker Recognition Evaluation Corpus

	[[category:Seminars]]		[[category:Seminars]]
			[[category:Seminars 2006]]

David: /* Speaker */

2006-06-12T17:33:59Z

Speaker

← Older revision		Revision as of 17:33, 12 June 2006
Line 6:		Line 6:
	== Speaker ==		== Speaker ==

	* [[Hugo Cordeiro]]		* Hugo Cordeiro
	* [[Carlos Meneses Ribeiro]]		* Carlos Meneses Ribeiro

	== Abstract ==		== Abstract ==

Lcoheur at 19:42, 8 June 2006

2006-06-08T19:42:44Z

New page

__NOTOC__
== Date ==

* June 16, 2006

== Speaker ==

* [[Hugo Cordeiro]]
* [[Carlos Meneses Ribeiro]]

== Abstract ==

The work described in this paper concerns the analysis of an alternative feature for speaker characterization, in the context of speaker recognition: Line Spectrum Frequencies (LSF), but derived from mel-filter bank energies. This new feature, that we denominate mel-LSFs (MLSFs), shows similar performance comparing to MFCCs for male speakers, one of the most common feature found in speaker recognition, but for female speakers MLSFs performs better than MFCCs.
When combined with mel LSFs differences, MLSFs feature overcomes the performance of the MFCCs for male and female speakers, even with temporal delta, ΔMFCCs, included. Performance is measured in the context of speaker verification, using EER and minimum HTER. Detection error threshold (DET) curves are also presented, as well as HTER curves.
The main objective of this study is to compare different features performances with a common framework, from what a standard support vector machine recogniser was developed.
Tests are based on the cellular component of the “2002 NIST Speaker Recognition Evaluation Corpus
[[category:Seminars]]