Open AccessProceedings Article
Frame pruning for automatic speaker identification
L. Besacier,J.F. Bonastre +1 more
- pp 1-4
TLDR
Validation of the pruning procedure on 567 speakers leads to a significative improvement on TIMIT and NTIMIT (up to 30% error rate reduction on TIM IT) and a prior frame level likelihood normalization in order to make comparison between frames meaningful.Abstract:
In this paper, we propose a frame selection procedure for text-independent speaker identification Instead of averaging the frame likelihoods along the whole test utterance, some of these are rejected (pruning) and the final score is computed with a limited number of frames This pruning stage requires a prior frame level likelihood normalization in order to make comparison between frames meaningful This normalization procedure alone leads to a significative performance enhancement As far as pruning is concerned, the optimal number of frames pruned is learned on a tuning data set for normal and telephone speech Validation of the pruning procedure on 567 speakers leads to a significative improvement on TIMIT and NTIMIT (up to 30% error rate reduction on TIMIT)read more
Citations
More filters
On the use of score pruning in speaker verification for speaker dependent threshold estimation.
Javier R. Saeta,Javier Hernando +1 more
TL;DR: Before estimating the threshold, score pruning removes outliers and improves subsequent estimations, and to solve the problem of impostor data, a speaker dependent threshold estimation with only data from clients is suggested.
Proceedings ArticleDOI
Reliable Speaker Identification Using Multiple Microphones in Ubiquitous Robot Companion Environment
TL;DR: The speaker identification system can provide human-robot interaction with a reliable basic interface with high classification accuracy and improve the recognition performance of speaker identification with multiple microphones on the robot side in adverse distant-talking environments.
Decisión threshold estimation and model quality evaluation techniques for speaker verification.
TL;DR: In this article, a tesis doctoral se centra en las etapas de entrenamiento and decision of un sistema de verificacion de locutores.
Journal ArticleDOI
The use of adaptive frame for speech recognition
Sam Kwong,Qianhua He +1 more
TL;DR: Word recognition experiments on the TIMIT and NON-TIMIT with discrete Hidden Markov Model (HMM) and continuous density HMM showed that steady performance improvement could be achieved for open set testing, proving the effectiveness of the proposed adaptive frame length feature extraction scheme especially for the open testing.
Proceedings ArticleDOI
Text-Independent Speaker Iden tification using Soft Channel Selection in a Multi-Microphone Environment
TL;DR: In this article, a text-independent speaker identification system was proposed to improve speaker identification in a multi-microphone environment, which incorporates soft channel selection before the combination of the identification results obtained by multiple microphones.
References
More filters
Journal ArticleDOI
Statistical Pattern Recognition
TL;DR: Introduction to statistical pattern recognition and nonlinear discriminant analysis - statistical methods.
Journal ArticleDOI
Text-independent speaker identification
H. Gish,M. Schmidt +1 more
TL;DR: A robust speaker-identification system is presented that was able to deal with various forms of anomalies that are localized in time, such as spurious noise events and crosstalk.
Proceedings ArticleDOI
NTIMIT: a phonetically balanced, continuous speech, telephone bandwidth speech database
TL;DR: The creation of the network TIMIT (NTIMIT) database, which is the result of transmitting the TIMIT database over the telephone network, is described, including characteristics useful for speech analysis and recognition.
Journal ArticleDOI
Second-order statistical measures for text-independent speaker identification
TL;DR: The use of some of the proposed measures as a reference benchmark to evaluate the intrinsic complexity of a given database under a given protocol is suggested as a conclusion to this work.
Related Papers (5)
Normalizations and selection of speech segments for speaker recognition scoring
K.-P. Li,J.E. Porter +1 more