Willkommen, schön sind Sie da!
Logo Ex Libris

Robust Adaptation to Non-Native Accents in Automatic Speech Recognition

  • E-Book (pdf)
  • 146 Seiten
(0) Erste Bewertung abgeben
Bewertungen
(0)
(0)
(0)
(0)
(0)
Alle Bewertungen ansehen
Speech recognition technology is being increasingly employed in human-machine interfaces. A remaining problem however is the robu... Weiterlesen
CHF 71.90
Download steht sofort bereit
Informationen zu E-Books
E-Books eignen sich auch für mobile Geräte (sehen Sie dazu die Anleitungen).
E-Books von Ex Libris sind mit Adobe DRM kopiergeschützt: Erfahren Sie mehr.
Weitere Informationen finden Sie hier.
Bestellung & Lieferung in eine Filiale möglich

Beschreibung

Speech recognition technology is being increasingly employed in human-machine interfaces. A remaining problem however is the robustness of this technology to non-native accents, which still cause considerable difficulties for current systems. In this book, methods to overcome this problem are described. A speaker adaptation algorithm that is capable of adapting to the current speaker with just a few words of speaker-specific data based on the MLLR principle is developed and combined with confidence measures that focus on phone durations as well as on acoustic features. Furthermore, a specific pronunciation modelling technique that allows the automatic derivation of non-native pronunciations without using non-native data is described and combined with the previous techniques to produce a robust adaptation to non-native accents in an automatic speech recognition system.



Inhalt

ASR:AnOverview.- Pre-processing of the Speech Data.- Stochastic Modelling of Speech.- Knowledge Bases of an ASR System.- Speaker Adaptation.- Confidence Measures.- Pronunciation Adaptation.- Future Work.- Summary.- Databases and Experimental Settings.- MLLR Results.- Phoneme Inventory.

Produktinformationen

Titel: Robust Adaptation to Non-Native Accents in Automatic Speech Recognition
Autor:
EAN: 9783540362906
Format: E-Book (pdf)
Hersteller: Springer Berlin Heidelberg
Genre: IT & Internet
Veröffentlichung: 01.07.2003
Digitaler Kopierschutz: Wasserzeichen
Anzahl Seiten: 146

Weitere Bände aus der Buchreihe "Lecture Notes in Computer Science"