A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2017; you can also visit the original URL.
The file type is application/pdf.
Filters
Cross-Language Spoken Document Retrieval on the TREC SDR Collection
[chapter]
2003
Lecture Notes in Computer Science
Preserved Fulltext
The benchmark is based on resources used in the last two spoken document retrieval tracks at the TREC conference, which are available on the Internet. ...
The extension from monolingual to cross-language SDR was obtained by translating all topics into five European languages: Dutch, French, German, Italian, and Spanish. ...
Acknowledgements This work was carried out within the project WebFAQ funded under the FDR-PAT program of the Province of Trento. ...
doi:10.1007/978-3-540-45237-9_41
fatcat:5auavwilavaz7naoclcxyjty54
The Cambridge University spoken document retrieval system
1999
1999 IEEE International Conference on Acoustics, Speech, and Signal Processing. Proceedings. ICASSP99 (Cat. No.99CH36258)
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2017; you can also visit the original URL.
The file type is application/pdf.
The results demonstrate the importance of high accuracy automatic transcription. The final system is currently being evaluated on the 1998 TREC-7 spoken document retrieval task. IEEE 1999 ...
The recognition engine is based on the HTK broadcast news transcription system and the retrieval engine is based on the techniques developed at City University. ...
Acknowledgements Thanks to Steve Renals for access to Sheffield University's TREC-6 SDR transcriptions. ...
doi:10.1109/icassp.1999.758059
dblp:conf/icassp/JohnsonJMJW99
fatcat:63szfri5mnhcdld6w6d26qlque
Automatic Processing Of Broadcast Audio In Multiple Languages
2002
Zenodo
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Publication in the conference proceedings of EUSIPCO, Toulouse, France, 2002 ...
The authors thank their colleagues in the Spoken Language Processing group at LIMSI for their participation in the development of different aspects of the automatic transcription and indexation systems ...
ACKNOWLEDGMENTS This work has been partially financed by the European Commission and the French Ministry of Defense. ...
doi:10.5281/zenodo.53630
fatcat:c6cghksa2jg4lgvlzvndmljs5a
Structuring Broadcast Audio for Information Access
2003
EURASIP Journal on Advances in Signal Processing
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2017; you can also visit the original URL.
The file type is application/pdf.
At Laboratoire d'Informatique pour la Mécanique et les Sciences de l'Ingénieur (LIMSI), broadcast news transcription systems have been developed for seven languages: English, French, German, Mandarin, ...
The transcription systems have been integrated into prototype demonstrators for several application areas such as audio data mining, structuring audiovisual archives, selective dissemination of information ...
The authors thank their colleagues in the Spoken Language Processing Group at LIMSI for their participation in the development of different aspects of the automatic transcription and indexing systems reported ...
doi:10.1155/s1110865703211033
fatcat:uouvssghczfrvbzekvj5chk32y
Spoken document representations for probabilistic retrieval
2000
Speech Communication
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
The overall improvement of the retrieval system can also be observed for seven dierent sets of transcriptions from dierent recognition engines with a WER ranging from 24.8% to 61.5%. ...
Taken together, these techniques can improve Average Precision by over 19% relative to a system similar to that which we presented at TREC-7. ...
Retrieval systems
Baseline system (BL) Our current SDR baseline system, BL, uses most of the strategies applied in our TREC-7 SDR evaluation system. ...
doi:10.1016/s0167-6393(00)00021-2
fatcat:46h5apxjgjazrcrtduhubu3tyi
From Text Summarisation to Style-Specific Summarisation for Broadcast News
[chapter]
2004
Lecture Notes in Computer Science
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2008; you can also visit the original URL.
The file type is application/pdf.
We show that the portability of traditional text summarisation features to broadcast news is dependent on the diffusiveness of the information in the broadcast news story. ...
An analysis of two categories of news stories (containing only read speech or including some spontaneous speech) demonstrates the importance of the style and the quality of the transcript, when extracting ...
The cuhtk-s1 system in the 1999 TREC-8 SDR evaluation. 8 The shef-s1 system in the 1999 TREC-8 SDR evaluation. ...
doi:10.1007/978-3-540-24752-4_17
fatcat:lkyidhv2vvhozmowhkm3anmaiy
Multimedia information seeking through search and hyperlinking
2013
Proceedings of the 3rd ACM conference on International conference on multimedia retrieval - ICMR '13
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2017; you can also visit the original URL.
The file type is application/pdf.
The search test queries and link assessment for this task was generated using the Amazon Mechanical Turk crowdsourcing platform. ...
The results of our experiments are used to propose a research agenda for developing effective techniques for search and hyperlinking of multimedia content. ...
The spoken document retrieval (SDR) task at TREC [8] required participants to find relevant audio recordings based on textual queries. ...
doi:10.1145/2461466.2461511
dblp:conf/mir/EskevichJAOCNGGSNDWGPL13
fatcat:3j22qwz2prguhm24cn647kdmkm
Searching spontaneous conversational speech
2007
SIGIR Forum
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2017; you can also visit the original URL.
The file type is application/pdf.
(LVCSR) systems could be built for the planned speech of news announcers. ...
Preface Nearly a decade ago, we learned from the TREC Spoken Document Retrieval (SDR) track that searching speech was a "solved problem." ...
The authors would like to thank Matthias Zimmermann, Yang Liu, and Mathew Magimai Doss for their help and suggestions. ...
doi:10.1145/1328964.1328982
fatcat:wwpzqq7ndrfedh4imhoznvccl4
Automatic speech recognition and its application to information extraction
1999
Proceedings of the 37th annual meeting of the Association for Computational Linguistics on Computational Linguistics -
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2005; you can also visit the original URL.
The file type is application/pdf.
This paper describes recent progress and the author's perspectives of speech recognition technology. ...
Applications of speech recognition technology can be classified into two main areas, dictation and human-computer dialogue systems. • ...
In the DARPA project, the Spoken Document Retrieval (SDR) of TREC and the Topic Detection and Tracking (TDT) program are supported by the same materials and systems that have been developed in the broadcast ...
doi:10.3115/1034678.1034680
dblp:conf/acl/Furu99
fatcat:ovnd2rgnwrf5ddhnnshwtppxje
Automatic tagging and geotagging in video collections and communities
2011
Proceedings of the 1st ACM International Conference on Multimedia Retrieval - ICMR '11
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2017; you can also visit the original URL.
The file type is application/pdf.
We overview three tasks offered in the MediaEval 2010 benchmarking initiative, for each, describing its use scenario, definition and the data set released. ...
For each task, a reference algorithm is presented that was used within MediaEval 2010 and comments are included on lessons learned. ...
The first major spoken-content-based benchmark, TREC Spoken Document Retrieval (TREC-SDR) [12] was devoted broadcast news retrieval and ran from 1997-2000. ...
doi:10.1145/1991996.1992047
dblp:conf/mir/LarsonSS11
fatcat:64ojdcpp3rda3bdt5zfl3knura
The THISL SDR system at TREC-8
2017
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
This paper describes the participation of the THISL group at the TREC-8 Spoken Document Retrieval (SDR) track. ...
The TREC-8 evaluation assessed SDR performance on a corpus of 500 hours of broadcast news material collected over a five month period. ...
The TREC-8 SDR track was designed to test how SDR systems perform with a much larger document collection than they have been evaluated on previously -the TREC-7 SDR track used only 87 hours of broadcast ...
doi:10.7916/d8fj2s36
fatcat:crz5utpn6nhq3g7gtc33dyzkga
Effect of Segmentation Method on Video Retrieval Performance
2005 IEEE International Conference on Multimedia and Expo
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2017; you can also visit the original URL.
The file type is application/pdf.
Moreover, in the case where manually segmented data are available for training, the approach combining the different modalities can lead to IR results close to those obtained with a manual segmentation ...
The results suggest that even with the sliding window segmentation, acceptable performance can be obtained on a broadcast news retrieval task. ...
For a more complete evaluation, we also used a second method (initially introduced in the context of TREC SDR [4] ) in which a system that outputs a ranking of time pointers is evaluated. ...
doi:10.1109/icme.2005.1521346
dblp:conf/icmcs/GrangierV05
fatcat:lvnosnzajnhopaye2f6jb5evfm
Are extractive text summarisation techniques portable to broadcast news?
2003 IEEE Workshop on Automatic Speech Recognition and Understanding (IEEE Cat. No.03EX721)
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2017; you can also visit the original URL.
The file type is application/pdf.
However, the quality of the speech transcripts as well as the difference in information structure between broadcast and newspaper news affect the usability of the individual features. ...
In this paper we report on a series of experiments which compare the effect of individual features on both text and speech summarisation, the effect of basing the speech summaries on automatic speech recognition ...
Figure 3 shows the ROC curves for speech summarisers based on transcripts from six different ASR systems (produced for the TREC-8 SDR evaluation), along with the manual transcript. ...
doi:10.1109/asru.2003.1318489
fatcat:pggbxepnrrcilat3dccbweler4
Spontaneous speech and opinion detection: mining call-centre transcripts
2013
Language Resources and Evaluation
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2017; you can also visit the original URL.
The file type is application/pdf.
), or the TREC 7-Spoken Document Retrieval, SDR-(Garofolo et al. 1999). ...
The F-scores obtained are 0.79 for business concepts detection, 0.74 for opinion detection and 0.67 for the extraction of relations between opinions and their target. ...
Acknowledgments This work was partly financed by CAP DIGITAL, the Business Cluster for digital content through the VoxFactory project. ...
doi:10.1007/s10579-013-9224-5
fatcat:jwshqp6utngwnmbnvsiw5k5va4
Investigating different models for cross-language information retrieval from automatic speech transcripts
2013
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
The availability of open source IR systems make it possible for us to investigate different Information Retrieval techniques, which proved their effectiveness in the literature for text retrieval, but ...
An ASR system is first used to transcribe digitized audio into text, and then a text retrieval system is used to retrieve speech segments, given a user request (information need). ...
SDR evaluation was implemented in 1999 for TREC-8 [1] . ...
doi:10.20381/ruor-13199
fatcat:ksd2cgasnnf4db5k2pvv22yyyy