Cross-lingual study of ASR errors: on the role of the context in human perception of near homophones - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2011

Cross-lingual study of ASR errors: on the role of the context in human perception of near homophones

Résumé

It is widely acknowledged that human listeners significantly outperform machines when it comes to transcribing speech. This paper presents a paradigm for perceptual experiments that aims to increase our understanding of human and automatic speech recognition errors. The role of the context length is investigated through perceptual recovery of small homophonic words or near-homophones yielding frequent automatic transcription errors. The same experimental protocol of varied size speech stimuli transcription is applied to both French and English. Our hypothesis is that ambiguity due to homophonic words reduces with context size for both languages, which in turn should entail reduced perception and transcription errors. The results show that context plays a central role as the human word error rate decreases significantly with increasing context. The long-term aim is to improve the modelling of such ambiguous items to reduce automatic errors.
Fichier non déposé

Dates et versions

hal-01135150 , version 1 (24-03-2015)

Identifiants

  • HAL Id : hal-01135150 , version 1

Citer

Ioana Vasilescu, D. Yahia, Natalie D. Snoeren, Martine Adda-Decker, Lori Lamel. Cross-lingual study of ASR errors: on the role of the context in human perception of near homophones. 12th Annual Conference of the International Speech Communication Association (Interspeech'11), International Speech Communication Association (ISCA), Aug 2011, Florence, Italy. pp.1949--1952. ⟨hal-01135150⟩
137 Consultations
0 Téléchargements

Partager

Gmail Facebook X LinkedIn More