Lexicon optimization for Dutch speech recognition in spoken document retrieval

Roeland J.F. Ordelman; Adrianus J. van Hessen; Franciska M.G. de Jong

Lexicon optimization for Dutch speech recognition in spoken document retrieval

Roeland J.F. Ordelman, Adrianus J. van Hessen, Franciska M.G. de Jong

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Professional

75 Downloads (Pure)

Abstract

In this paper, ongoing work concerning the language modelling and lexicon optimization of a Dutch speech recognition system for Spoken Document Retrieval is described: the collection and normalization of a training data set and the optimization of our recognition lexicon. Effects on lexical coverage of the amount of training data, of decompounding compound words and of different selection methods for proper names and acronyms are discussed.

Original language	English
Title of host publication	Proceedings of Eurospeech 2001 - Scandinavia
Editors	P. Dalsgaard, B. Lindberg, H. Benner
Pages	1085-1088
Publication status	Published - 2001
Event	Conferentie in Aalborg, Denmark: Proceedings of Eurospeech 2001 - Scandinavia - Duration: 1 Jan 1900 → …

Conference

Conference	Conferentie in Aalborg, Denmark
Period	1/01/00 → …

Access to Document

ArticleFinal published version, 69.2 KB

Cite this

@inproceedings{8ead7d5ecc6d4d4abbbd2f5313e03b14,

title = "Lexicon optimization for Dutch speech recognition in spoken document retrieval",

abstract = "In this paper, ongoing work concerning the language modelling and lexicon optimization of a Dutch speech recognition system for Spoken Document Retrieval is described: the collection and normalization of a training data set and the optimization of our recognition lexicon. Effects on lexical coverage of the amount of training data, of decompounding compound words and of different selection methods for proper names and acronyms are discussed.",

author = "Ordelman, {Roeland J.F.} and {van Hessen}, {Adrianus J.} and {de Jong}, {Franciska M.G.}",

year = "2001",

language = "English",

isbn = "87-90834-09-7",

pages = "1085--1088",

editor = "P. Dalsgaard and B. Lindberg and H. Benner",

booktitle = "Proceedings of Eurospeech 2001 - Scandinavia",

note = "Conferentie in Aalborg, Denmark : Proceedings of Eurospeech 2001 - Scandinavia ; Conference date: 01-01-1900",

}

Lexicon optimization for Dutch speech recognition in spoken document retrieval. / Ordelman, Roeland J.F.; van Hessen, Adrianus J.; de Jong, Franciska M.G.
Proceedings of Eurospeech 2001 - Scandinavia. ed. / P. Dalsgaard; B. Lindberg; H. Benner. 2001. p. 1085-1088.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Professional

TY - GEN

T1 - Lexicon optimization for Dutch speech recognition in spoken document retrieval

AU - Ordelman, Roeland J.F.

AU - van Hessen, Adrianus J.

AU - de Jong, Franciska M.G.

PY - 2001

Y1 - 2001

N2 - In this paper, ongoing work concerning the language modelling and lexicon optimization of a Dutch speech recognition system for Spoken Document Retrieval is described: the collection and normalization of a training data set and the optimization of our recognition lexicon. Effects on lexical coverage of the amount of training data, of decompounding compound words and of different selection methods for proper names and acronyms are discussed.

AB - In this paper, ongoing work concerning the language modelling and lexicon optimization of a Dutch speech recognition system for Spoken Document Retrieval is described: the collection and normalization of a training data set and the optimization of our recognition lexicon. Effects on lexical coverage of the amount of training data, of decompounding compound words and of different selection methods for proper names and acronyms are discussed.

M3 - Conference contribution

SN - 87-90834-09-7

SP - 1085

EP - 1088

BT - Proceedings of Eurospeech 2001 - Scandinavia

A2 - Dalsgaard, P.

A2 - Lindberg, B.

A2 - Benner, H.

T2 - Conferentie in Aalborg, Denmark

Y2 - 1 January 1900

ER -

Lexicon optimization for Dutch speech recognition in spoken document retrieval

Abstract

Conference

Access to Document

Fingerprint

Cite this