28604 articles – 22086 references  [version française]

inria-00100780, version 1

Reconnaissance de la parole pour les locuteurs non natifs en présence de bruit

Dominique Fohr () a1, Odile Mella b1, Irina Illina c1, Fabrice Lauri d1, Christophe Cerisara () a1, Christophe Antoine d1

XXIVèmes Journées d'Etude sur la Parole - JEP'02 (2002) 297-301

Abstract: In real world applications, speech recognition is confronted with two main difficulties: the non native speakers and the background noise. The aim of this paper is to compare on the same noisy database different methods in order to increase the robustness of our HMM-based automatic speech recognition system. To deal with the non native speakers, we have tested two solutions: multi-models and adaptation techniques. For noisy speech, we have evaluated two types of methods: the first one (PMC and MLLR) adapts the initial models, trained in clean speech, with a few noisy sentences. The second one (RATZ and MCR) tries to remove the noise from the signal without modifying the HMM models.

  • a –  CNRS
  • b –  UNIVERSITE HENRI POINCARE
  • c –  UNIVERSITE NANCY 2
  • d –  INRIA
  • 1:  PAROLE (INRIA Lorraine - LORIA)
  • INRIA – CNRS : UMR7503 – Université Henri Poincaré - Nancy I – Université Nancy II – Institut National Polytechnique de Lorraine (INPL)
  • Domain : Computer Science/Other
  • Keywords : automatic speech recognition – non native speakers – background noise || reconnaissance automatique de la parole – parole bruitée – locuteurs non natives
  • Internal note : A02-R-112 || fohr02a
  • Comment : Colloque avec actes et comité de lecture. nationale.
 
  • inria-00100780, version 1
  • oai:hal.inria.fr:inria-00100780
  • From: 
  • Submitted on: Tuesday, 26 September 2006 14:50:59
  • Updated on: Thursday, 28 September 2006 15:22:47