Patrol team language identification system for DARPA RATS P1 evaluation

Matějka, Pavel and Plchot, Oldřich and Soufifar, Mehdi and Glembek, Ondřej and D'haro Enríquez, Luis Fernando and Veselý, Karel and Grézl, František and Ma, Jeff and Matsoukas, Spyros and Dehak, Najim (2012). Patrol team language identification system for DARPA RATS P1 evaluation. In: "InterSpeech 2012 - 13th Annual Conference of the International Speech Communication Association", 09/09/2012 - 13/09/2012, Portland, Oregon. pp. 1-4.

Description

Title: Patrol team language identification system for DARPA RATS P1 evaluation
Author/s:
  • Matějka, Pavel
  • Plchot, Oldřich
  • Soufifar, Mehdi
  • Glembek, Ondřej
  • D'haro Enríquez, Luis Fernando
  • Veselý, Karel
  • Grézl, František
  • Ma, Jeff
  • Matsoukas, Spyros
  • Dehak, Najim
Item Type: Presentation at Congress or Conference (Article)
Event Title: InterSpeech 2012 - 13th Annual Conference of the International Speech Communication Association
Event Dates: 09/09/2012 - 13/09/2012
Event Location: Portland, Oregon
Title of Book: InterSpeech 2012 - 13th Annual Conference of the International Speech Communication Association
Date: 2012
Subjects:
Faculty: E.T.S.I. Telecomunicación (UPM)
Department: Ingeniería Electrónica
Creative Commons Licenses: Recognition - No derivative works - Non commercial

Full text

[img]
Preview
PDF - Requires a PDF viewer, such as GSview, Xpdf or Adobe Acrobat Reader
Download (64kB) | Preview

Abstract

This paper describes the language identification (LID) system developed by the Patrol team for the first phase of the DARPA RATS (Robust Automatic Transcription of Speech) program, which seeks to advance state of the art detection capabilities on audio from highly degraded communication channels. We show that techniques originally developed for LID on telephone speech (e.g., for the NIST language recognition evaluations) remain effective on the noisy RATS data, provided that careful consideration is applied when designing the training and development sets. In addition, we show significant improvements from the use of Wiener filtering, neural network based and language dependent i-vector modeling, and fusion.

More information

Item ID: 20384
DC Identifier: http://oa.upm.es/20384/
OAI Identifier: oai:oa.upm.es:20384
Deposited by: Memoria Investigacion
Deposited on: 05 Oct 2013 07:35
Last Modified: 21 Apr 2016 23:08
  • Logo InvestigaM (UPM)
  • Logo GEOUP4
  • Logo Open Access
  • Open Access
  • Logo Sherpa/Romeo
    Check whether the anglo-saxon journal in which you have published an article allows you to also publish it under open access.
  • Logo Dulcinea
    Check whether the spanish journal in which you have published an article allows you to also publish it under open access.
  • Logo de Recolecta
  • Logo del Observatorio I+D+i UPM
  • Logo de OpenCourseWare UPM