Speaker identification based on the time-delay Hierarchical Mixture of Experts

Ke Chen; Dahong Xie; Huisheng Chi

Speaker identification based on the time-delay Hierarchical Mixture of Experts

Ke Chen, Dahong Xie, Huisheng Chi

Research output: Chapter in Book/Conference proceeding › Conference contribution › peer-review

Abstract

In this paper, we explore the Hierarchical Mixture of Experts (HME) architecture for a substantial problem, that of text-dependent speaker identification. For a specific multi-way classification, we propose a generalized Bernolli density instead of the multinomial logit density. Time-delay technique is also introduced to HME for spatio-temporal processing. Using the proposed density and the time-delay HME along with the EM algorithm, we show that the system has a satisfactory performance and yields significantly fast training.

Original language	English
Title of host publication	IEEE International Conference on Neural Networks - Conference Proceedings\|IEEE Int Conf Neural Networks Conf Proc
Place of Publication	Piscataway, NJ, United States
Publisher	IEEE
Pages	2062-2066
Number of pages	4
Volume	4
Publication status	Published - 1995
Event	Proceedings of the 1995 IEEE International Conference on Neural Networks. Part 1 (of 6) - Perth, Aust Duration: 1 Jul 1995 → …

Conference

Conference	Proceedings of the 1995 IEEE International Conference on Neural Networks. Part 1 (of 6)
City	Perth, Aust
Period	1/07/95 → …

Cite this

@inproceedings{75c275fcbf4f4643a672c1179e593da8,

title = "Speaker identification based on the time-delay Hierarchical Mixture of Experts",

abstract = "In this paper, we explore the Hierarchical Mixture of Experts (HME) architecture for a substantial problem, that of text-dependent speaker identification. For a specific multi-way classification, we propose a generalized Bernolli density instead of the multinomial logit density. Time-delay technique is also introduced to HME for spatio-temporal processing. Using the proposed density and the time-delay HME along with the EM algorithm, we show that the system has a satisfactory performance and yields significantly fast training.",

author = "Ke Chen and Dahong Xie and Huisheng Chi",

year = "1995",

language = "English",

volume = "4",

pages = "2062--2066",

booktitle = "IEEE International Conference on Neural Networks - Conference Proceedings|IEEE Int Conf Neural Networks Conf Proc",

publisher = "IEEE",

address = "United States",

note = "Proceedings of the 1995 IEEE International Conference on Neural Networks. Part 1 (of 6) ; Conference date: 01-07-1995",

}

Chen, K, Xie, D & Chi, H 1995, Speaker identification based on the time-delay Hierarchical Mixture of Experts. in IEEE International Conference on Neural Networks - Conference Proceedings|IEEE Int Conf Neural Networks Conf Proc. vol. 4, IEEE, Piscataway, NJ, United States, pp. 2062-2066, Proceedings of the 1995 IEEE International Conference on Neural Networks. Part 1 (of 6), Perth, Aust, 1/07/95.

Speaker identification based on the time-delay Hierarchical Mixture of Experts. / Chen, Ke; Xie, Dahong; Chi, Huisheng.
IEEE International Conference on Neural Networks - Conference Proceedings|IEEE Int Conf Neural Networks Conf Proc. Vol. 4 Piscataway, NJ, United States: IEEE, 1995. p. 2062-2066.

Research output: Chapter in Book/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Speaker identification based on the time-delay Hierarchical Mixture of Experts

AU - Chen, Ke

AU - Xie, Dahong

AU - Chi, Huisheng

PY - 1995

Y1 - 1995

N2 - In this paper, we explore the Hierarchical Mixture of Experts (HME) architecture for a substantial problem, that of text-dependent speaker identification. For a specific multi-way classification, we propose a generalized Bernolli density instead of the multinomial logit density. Time-delay technique is also introduced to HME for spatio-temporal processing. Using the proposed density and the time-delay HME along with the EM algorithm, we show that the system has a satisfactory performance and yields significantly fast training.

AB - In this paper, we explore the Hierarchical Mixture of Experts (HME) architecture for a substantial problem, that of text-dependent speaker identification. For a specific multi-way classification, we propose a generalized Bernolli density instead of the multinomial logit density. Time-delay technique is also introduced to HME for spatio-temporal processing. Using the proposed density and the time-delay HME along with the EM algorithm, we show that the system has a satisfactory performance and yields significantly fast training.

UR - https://www.scopus.com/pages/publications/0029519886

M3 - Conference contribution

VL - 4

SP - 2062

EP - 2066

BT - IEEE International Conference on Neural Networks - Conference Proceedings|IEEE Int Conf Neural Networks Conf Proc

PB - IEEE

CY - Piscataway, NJ, United States

T2 - Proceedings of the 1995 IEEE International Conference on Neural Networks. Part 1 (of 6)

Y2 - 1 July 1995

ER -

Speaker identification based on the time-delay Hierarchical Mixture of Experts

Abstract

Conference

Other files and links

Fingerprint

Cite this