Room Localization for Distant Speech Recognition

Juan Andrés Morales Cordovilla, Hannes Pessentheiner, Martin Hagmüller, Gernot Kubin

Publikation: Beitrag in Buch/Bericht/KonferenzbandBeitrag in einem KonferenzbandBegutachtung

Abstract

The problem of room localization is to determine where, in a multi-room environment, a person is producing a speech utterance. In our work, we are exploiting the information gained
from a network of microphones installed all over a house, where the lack of calibration of the microphone energies creates an additional challenge. This paper compares room localizers based on different features (such as energy and cross-correlation between microphones) and classifiers (such as neural networks and discriminative analysis). In order to evaluate the different room localizers in terms of word accuracy this paper also presents a complete distant speech recognition system which tries to take advantage of synergy between the different components without using any oracle information. Finally, the system is analyzed in terms of computational and time resources
Originalspracheenglisch
TitelCelebrating the diversity of spoken languages
Untertitel15th Annual Conference of the International Speech Communication Association
ErscheinungsortRed Hook, NY
Herausgeber (Verlag)Curran
Seiten2450-2453
Band4
ISBN (Print)9781634394352
PublikationsstatusVeröffentlicht - 2014
Veranstaltung15th International Conference on Spoken Language Processing: Interspeech 2014 - Singapur, Singapur
Dauer: 14 Sept. 201418 Sept. 2014

Konferenz

Konferenz15th International Conference on Spoken Language Processing
KurztitelInterspeech 2014
Land/GebietSingapur
OrtSingapur
Zeitraum14/09/1418/09/14

Fields of Expertise

  • Information, Communication & Computing

Treatment code (Nähere Zuordnung)

  • Basic - Fundamental (Grundlagenforschung)
  • Application

Fingerprint

Untersuchen Sie die Forschungsthemen von „Room Localization for Distant Speech Recognition“. Zusammen bilden sie einen einzigartigen Fingerprint.

Dieses zitieren