Building voice lineups: one method's bias problems

Auteurs-es

  • A.S. Laubstein Dept. Linguistics, Carleton Univ., Ottawa, Ont., Canada

Mots-clés :

speech recognition, voice recognition, voice lineups, bias, transcript method, speech content

Résumé

This paper describes an experimental study which investigates the feasibility of using what shall be termed the `transcript method; in the preparation of voice lineups-the currently accepted procedure in Canada and the United States (Mayor, 1989). This method has been used in an attempt to eliminate the speech content from the voice lineup. It was for instance, the method chosen by the Ottawa Police in a recent investigation. There are a number of problems with this method-in particular, it is not clear that it is capable of producing an unbiased lineup

Fichiers supplémentaires

Publié-e

1994-09-01

Comment citer

1.
Laubstein A. Building voice lineups: one method’s bias problems. Canadian Acoustics [Internet]. 1 sept. 1994 [cité 12 mai 2026];22(3):135-6. Disponible à: https://jcaa.caa-aca.ca/index.php/jcaa/article/view/886

Numéro

Rubrique

Actes du congrès de la Semaine canadienne d'acoustique