Decision trees with improved efficiency for fast speaker verification

Gilles Gonon 1 Rémi Gribonval 1 Frédéric Bimbot 1
1 METISS - Speech and sound data modeling and processing
IRISA - Institut de Recherche en Informatique et Systèmes Aléatoires, Inria Rennes – Bretagne Atlantique
Abstract : Classification and regression trees (CART) are convenient for low complexity speaker recognition on embedded devices. However, former attempts at using trees performed quite poorly compared to state of the art results with Gaussian Mixture Models (GMM). In this article, we introduce some solutions to improve the efficiency of the tree-based approach. First, we propose to use at the tree construction level different types of information from the GMM used in state of the art techniques. Then, we model the score function within each leaf of the tree by a linear score function. Considering a baseline state of the art system with an equal error rate (EER) of 8.6% on the NIST 2003 evaluation, a previous CART method provides typical EER ranging between 16% and 18% while the proposed improvements decrease the EER to 11.5%, with a computational cost suitable for embedded devices.
Complete list of metadatas

Cited literature [5 references]  Display  Hide  Download

https://hal.inria.fr/inria-00564509
Contributor : Rémi Gribonval <>
Submitted on : Wednesday, February 9, 2011 - 9:28:24 AM
Last modification on : Friday, November 16, 2018 - 1:22:35 AM
Long-term archiving on : Tuesday, May 10, 2011 - 2:57:19 AM

File

2005_Interspeech_GononGribBimb...
Files produced by the author(s)

Identifiers

  • HAL Id : inria-00564509, version 1

Citation

Gilles Gonon, Rémi Gribonval, Frédéric Bimbot. Decision trees with improved efficiency for fast speaker verification. 9th European Conference on Speech Communication and Technology, Proceedings of (INTERSPEECH-2005), Sep 2005, Lisbonne, Portugal. pp.3077--3080. ⟨inria-00564509⟩

Share

Metrics

Record views

441

Files downloads

254