Logo des Repositoriums
 

Speech recognition as a retrieval problem

dc.contributor.authorRieber, Joscha Simon
dc.contributor.authorBardeli, Rolf
dc.contributor.editorHorbach, Matthias
dc.date.accessioned2019-03-07T09:32:27Z
dc.date.available2019-03-07T09:32:27Z
dc.date.issued2013
dc.description.abstractCommon approaches to automatic speech recognition (ASR) are based on training statistical models for the acoustics of speech. In our work, a retrieval-based ASR system is developed that does not rely on training and thus provides more flexible application. It is based on a set of known reference word utterances for each possibly occurring word in a test string. A test word string is identified by finding the most similar reference for each word by using an approach based on dynamic time warping (DTW). The DTW variant suitable for recognizing strings of connected words is called level-building DTW, proposed by Myers and Rabiner in 1981. It is using a level-bylevel iteration to match each word in the test utterance with the most similar reference. In our work, an ASR system for connected digit recognition based on level-building DTW is developed, evaluated and compared with a state-of-the-art HMM recognizer.en
dc.identifier.isbn978-3-88579-614-5
dc.identifier.pissn1617-5468
dc.identifier.urihttps://dl.gi.de/handle/20.500.12116/20711
dc.language.isoen
dc.publisherGesellschaft für Informatik e.V.
dc.relation.ispartofINFORMATIK 2013 – Informatik angepasst an Mensch, Organisation und Umwelt
dc.relation.ispartofseriesLecture Notes in Informatics (LNI) - Proceedings, Volume P-220
dc.titleSpeech recognition as a retrieval problemen
dc.typeText/Conference Paper
gi.citation.endPage2971
gi.citation.publisherPlaceBonn
gi.citation.startPage2958
gi.conference.date16.-20. September 2013
gi.conference.locationKoblenz
gi.conference.sessiontitleRegular Research Papers

Dateien

Originalbündel
1 - 1 von 1
Lade...
Vorschaubild
Name:
2958.pdf
Größe:
343.6 KB
Format:
Adobe Portable Document Format