Automatic Language Model Adapation for Spoken Document Retrieval

Published

January 1, 2000

Author(s)

C G. Auzanne, John S. Garofolo, Jonathan G. Fiscus, W M. Fisher

Abstract

This paper describes experiments implemented at NIST in adapting language models over time to improve recognition of broadcast news recorded over many months. These experiments were designed specifically to improve the utility of automatically generated transcripts for retrieval applications. To evaluate the potential of the approach, a time-adaptive automatic speech recognition run was implemented to support the 1999 TREC Spoken Document Retrieval (SDR) Track - more than 500 hours of broadcast news sampled across 5 months. The accuracy of retrieval for several systems using the time-adaptive system transcripts was evaluated against transcripts produced by virtually the same recognition system with a fixed language model.This paper details the process we employed to identify and implement the time-adaptive language model and discusses the results of the experiment in terms of its effect on word error rate, out of vocabulary rate and retrieval accuracy (Mean Average Precision).

Proceedings Title

RIAO-2000 Content-Based Multimedia Information Access Conference

Conference Dates

January 1, 2000

Conference Location

Paris, FR

Conference Title

RIAO Conference on Content-Based Multimedia Information Access

Pub Type

Conferences

Download Paper

Local Download

Keywords

adaptation, broadcast, document, indexing multimedia, news, recognition, retrieval, spoken

Citation

Auzanne, C. , Garofolo, J. , Fiscus, J. and Fisher, W. (2000), Automatic Language Model Adapation for Spoken Document Retrieval, RIAO-2000 Content-Based Multimedia Information Access Conference, Paris, FR, [online], https://tsapps.nist.gov/publication/get_pdf.cfm?pub_id=151486 (Accessed October 8, 2024)