Castsearch - Context Based Spoken Document Retrieval

Lasse Lohilahti Mølgaard; Kasper Winther Jørgensen; Lars Kai Hansen

doi:10.1109/ICASSP.2007.367171

Castsearch - Context Based Spoken Document Retrieval

Lasse Lohilahti Mølgaard, Kasper Winther Jørgensen, Lars Kai Hansen

Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review

927 Downloads (Pure)

Abstract

The paper describes our work on the development of a system for retrieval of relevant stories from broadcast news. The system utilizes a combination of audio processing and text mining. The audio processing consists of a segmentation step that partitions the audio into speech and music. The speech is further segmented into speaker segments and then transcribed using an automatic speech recognition system, to yield text input for clustering using non-negative matrix factorization (NMF). We find semantic topics that are used to evaluate the performance for topic detection. Based on these topics we show that a novel query expansion can be performed to return more intelligent search results. We also show that the query expansion helps overcome errors of the automatic transcription

Original language	English
Title of host publication	IEEE International Conference on Acoustics, Speech and Signal Processing, 2007. ICASSP 2007.
Volume	4
Publisher	IEEE
Publication date	2007
ISBN (Print)	1-4244-0727-3
DOIs	https://doi.org/10.1109/ICASSP.2007.367171
Publication status	Published - 2007
Event	2007 IEEE International Conference on Acoustics, Speech and Signal Processing - Honolulu, United States Duration: 15 Apr 2007 → 20 Apr 2007 Conference number: 32

Conference

Conference	2007 IEEE International Conference on Acoustics, Speech and Signal Processing
Number	32
Country/Territory	United States
City	Honolulu
Period	15/04/2007 → 20/04/2007

Bibliographical note

Copyright: 2007 IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE

Keywords

Nonnegative Matrix Factorization
Audio Retrieval
Text Mining
Document Clustering

Access to Document

10.1109/ICASSP.2007.367171

MølgaardFinal published version, 138 KB
Imm4872

http://www.icassp2007.org/

OpenUrl availability

Full text

Cite this

@inproceedings{6673a87bebae4e56a83876eab773d580,

title = "Castsearch - Context Based Spoken Document Retrieval",

abstract = "The paper describes our work on the development of a system for retrieval of relevant stories from broadcast news. The system utilizes a combination of audio processing and text mining. The audio processing consists of a segmentation step that partitions the audio into speech and music. The speech is further segmented into speaker segments and then transcribed using an automatic speech recognition system, to yield text input for clustering using non-negative matrix factorization (NMF). We find semantic topics that are used to evaluate the performance for topic detection. Based on these topics we show that a novel query expansion can be performed to return more intelligent search results. We also show that the query expansion helps overcome errors of the automatic transcription",

keywords = "Nonnegative Matrix Factorization, Audio Retrieval, Text Mining, Document Clustering",

author = "M{\o}lgaard, {Lasse Lohilahti} and J{\o}rgensen, {Kasper Winther} and Hansen, {Lars Kai}",

note = "Copyright: 2007 IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE; 2007 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2007 ; Conference date: 15-04-2007 Through 20-04-2007",

year = "2007",

doi = "10.1109/ICASSP.2007.367171",

language = "English",

isbn = "1-4244-0727-3",

volume = "4",

booktitle = "IEEE International Conference on Acoustics, Speech and Signal Processing, 2007. ICASSP 2007.",

publisher = "IEEE",

address = "United States",

}

Mølgaard, LL, Jørgensen, KW & Hansen, LK 2007, Castsearch - Context Based Spoken Document Retrieval. in IEEE International Conference on Acoustics, Speech and Signal Processing, 2007. ICASSP 2007.. vol. 4, IEEE, 2007 IEEE International Conference on Acoustics, Speech and Signal Processing, Honolulu, Hawaii, United States, 15/04/2007. https://doi.org/10.1109/ICASSP.2007.367171

TY - GEN

T1 - Castsearch - Context Based Spoken Document Retrieval

AU - Mølgaard, Lasse Lohilahti

AU - Jørgensen, Kasper Winther

AU - Hansen, Lars Kai

N1 - Conference code: 32

PY - 2007

Y1 - 2007

N2 - The paper describes our work on the development of a system for retrieval of relevant stories from broadcast news. The system utilizes a combination of audio processing and text mining. The audio processing consists of a segmentation step that partitions the audio into speech and music. The speech is further segmented into speaker segments and then transcribed using an automatic speech recognition system, to yield text input for clustering using non-negative matrix factorization (NMF). We find semantic topics that are used to evaluate the performance for topic detection. Based on these topics we show that a novel query expansion can be performed to return more intelligent search results. We also show that the query expansion helps overcome errors of the automatic transcription

AB - The paper describes our work on the development of a system for retrieval of relevant stories from broadcast news. The system utilizes a combination of audio processing and text mining. The audio processing consists of a segmentation step that partitions the audio into speech and music. The speech is further segmented into speaker segments and then transcribed using an automatic speech recognition system, to yield text input for clustering using non-negative matrix factorization (NMF). We find semantic topics that are used to evaluate the performance for topic detection. Based on these topics we show that a novel query expansion can be performed to return more intelligent search results. We also show that the query expansion helps overcome errors of the automatic transcription

KW - Nonnegative Matrix Factorization

KW - Audio Retrieval

KW - Text Mining

KW - Document Clustering

U2 - 10.1109/ICASSP.2007.367171

DO - 10.1109/ICASSP.2007.367171

M3 - Article in proceedings

SN - 1-4244-0727-3

VL - 4

BT - IEEE International Conference on Acoustics, Speech and Signal Processing, 2007. ICASSP 2007.

PB - IEEE

T2 - 2007 IEEE International Conference on Acoustics, Speech and Signal Processing

Y2 - 15 April 2007 through 20 April 2007

ER -

Castsearch - Context Based Spoken Document Retrieval

Abstract

Conference

Bibliographical note

Keywords

Access to Document

OpenUrl availability

Fingerprint

Cite this