Filters








48,547 Hits in 2.4 sec

Searching spontaneous conversational speech

Franciska de Jong, Douglas W. Oard, Roeland Ordelman, Stephan Raaijmakers
2007 SIGIR Forum  
Preface Nearly a decade ago, we learned from the TREC Spoken Document Retrieval (SDR) track that searching speech was a "solved problem."  ...  . • The redundancy present in human language meant that search effectiveness held up well over a reasonable range of transcription accuracy. • Sufficiently accurate Large-Vocabulary Continuous Speech Recognition  ...  SEARCHING SPONTANEOUS SPEECH These observations potentially have significant implications for searching of spontaneous conversational as speech.  ... 
doi:10.1145/1328964.1328982 fatcat:wwpzqq7ndrfedh4imhoznvccl4

Searching multimedia content with a spontaneous conversational speech track

Martha Larson, Roeland Ordelman, Franciska de Jong, Wessel Kraaij, Joachim Kohler
2009 Proceedings of the seventeen ACM international conference on Multimedia - MM '09  
WHAT IS SPONTANEOUS CONVERSATIONAL SPEECH? We produce spontaneous conversational speech when we speak to each other without previously selecting words or composing sentences.  ...  Planned speech and spontaneous conversational speech also differ with respect to their overall structure.  ... 
doi:10.1145/1631272.1631549 dblp:conf/mm/LarsonOJKK09 fatcat:7tgumqkjljfmlgejqmgtvzdrdi

PodCastle

Jun Ogata, Masataka Goto
2009 Proceedings of the third workshop on Searching spontaneous conversational speech - SSCS '09  
We have developed a Web 2.0 service, PodCastle, that enables full-text searching of speech data (podcasts) on the basis of automatic speech recognition.  ...  PodCastle enables users to search and read podcasts, and to share the full text of speech recognition results for podcasts.  ...  by training our speech recognizer so that other podcasts can be searched more reliably.  ... 
doi:10.1145/1631127.1631133 fatcat:sssf4zrhgzeujgtoi5sv2tpjrm

Automatic indexing of speech segments with spontaneity levels on large audio database

Richard Dufour, Yannick Estève, Paul Deléglise
2010 Proceedings of the 2010 international workshop on Searching spontaneous conversational speech - SSCS '10  
Spontaneous speech detection from a large audio database can be useful for different applications.  ...  For example, processing spontaneous speech is one of the many challenges that Automatic Speech Recognition (ASR) systems have to deal with.  ...  speech, low spontaneity or high spontaneity label.  ... 
doi:10.1145/1878101.1878110 fatcat:62zq2qn7xbabnmuefbh7irs7ua

The ACLD

Andrei Popescu-Belis, Jonathan Kilgour, Alexandre Nanchen, Peter Poller
2010 Proceedings of the 2010 international workshop on Searching spontaneous conversational speech - SSCS '10  
The linked content is displayed in real-time to the participants in the conversation, or to users watching a recorded conversation or talk.  ...  The system can be demonstrated in both settings, using real-time automatic speech recognition (ASR) or replaying offline ASR, via a flexible user interface that displays results and provides access to  ...  The ASR can be coupled to a microphone array to improve recognition of conversational speech.  ... 
doi:10.1145/1878101.1878111 fatcat:m3q33f2sebb5fm7n7iqwl2fy7y

The ambient spotlight

Jonathan Kilgour, Jean Carletta, Steve Renals
2010 Proceedings of the 2010 international workshop on Searching spontaneous conversational speech - SSCS '10  
The prototype, which relies on meeting speech recognition and topic segmentation, formulates and runs desktop search queries in order to present its results.  ...  This interface is intended to supplement or replace the textual searches that managers typically perform.  ...  several attempts to formulate the right textual search string.  ... 
doi:10.1145/1878101.1878112 fatcat:nkt7yvzfefdabovh52llqyflt4

A parallel meeting diarist

Gerald Friedland, Jike Chong, Adam Janin
2010 Proceedings of the 2010 international workshop on Searching spontaneous conversational speech - SSCS '10  
This paper presents the application and the underlying parallel speaker diarization and speech recognition realizations.  ...  We therefore developed novel parallel methods for speaker diarization and speech recognition that are optimized to run on multicore and manycore architectures.  ...  The authors of [5] explored coarse-grained concurrency in large vocabulary conversational speech recognition (LVCSR) and implemented a pipeline of tasks on a cellphone-oriented multicore architecture  ... 
doi:10.1145/1878101.1878114 fatcat:5bam27jtmvf2vdmaobgzyv53eu

Large multimedia archive for world languages

Peter Wittenburg, Paul Trilsbeek, Przemek Lenkiewicz
2010 Proceedings of the 2010 international workshop on Searching spontaneous conversational speech - SSCS '10  
Due to an immediate conversion of the incoming data to standards-based formats and checks at upload time lifecycle management of all 50 Terabyte of data is widely simplified.  ...  Metadata selections can be used by these techniques which then can be used to carry out a content search via the TROVA [27] search engine.  ...  STANDARDS Since it is known that curation costs grow over time we apply an immediate conversion policy where possible.  ... 
doi:10.1145/1878101.1878113 fatcat:nooe4hblrng4ngle5bf6yw2emi

A latent semantic retrieval and clustering system for personal photos with sparse speech annotation

Yi-Sheng Fu, Winston H. Hsu, Lin-Shan Lee
2009 Proceedings of the third workshop on Searching spontaneous conversational speech - SSCS '09  
In this demo we present a user-friendly latent semantic retrieval and clustering system for personal photos with sparse spontaneous speech tags annotated when the photos were taken.  ...  Only 10% of the photos need to be annotated by spontaneous speech of a few words regarding one or two semantic categories (e.g. what or where), while all photos can be effectively retrieved using highlevel  ...  In this paper we will exploit the use of visual words with spontaneous speech and investigate feasibility for search result clustering [5] .  ... 
doi:10.1145/1631127.1631134 fatcat:faza5sajiba4bkqcr5vm5a6avy

Spoken news queries over the world wide web

Sebastian Stüker, Michael Heck, Katja Renner, Alex Waibel
2010 Proceedings of the 2010 international workshop on Searching spontaneous conversational speech - SSCS '10  
The client side of our architecture only requires a web browser with Flash extension in order to record and send the speech of the queries to the servers and in order to display the retrieved news clips  ...  Usually, search engines require the user to type a query, usually in the form of relevant keywords and phrases, in order to start the search process.  ...  But searching through multimedia content needs more processing, in order to make the unstructured video and audio streams suitable for current search technologies.  ... 
doi:10.1145/1878101.1878115 fatcat:ohxkt25yebf3diudx6qteqd5hi

Topic modeling for spoken document retrieval using word- and syllable-level information

Shih-Hsiang Lin, Berlin Chen
2009 Proceedings of the third workshop on Searching spontaneous conversational speech - SSCS '09  
Although most of the above approaches can be equally applied to both text and spoken documents, the latter presents unique difficulties, such as speech recognition errors, problems posed by spontaneous  ...  speech, or redundant information.  ... 
doi:10.1145/1631127.1631129 fatcat:ipf5nyow45b2pedpsofakgypd4

Direct posterior confidence for out-of-vocabulary spoken term detection

Dong Wang, Simon King, Nicholas Evans, Joe Frankel, Raphaél Troncy
2010 Proceedings of the 2010 international workshop on Searching spontaneous conversational speech - SSCS '10  
Our experiments, set up on multi-party meeting speech which is highly spontaneous and conversational, demonstrate that the proposed technique improves STD performance on OOV terms significantly; when combined  ...  Compared to conventional speech transcription and keyword spotting, STD is an open-vocabulary task and is necessarily required to address out-of-vocabulary (OOV) terms.  ...  Meeting speech is highly 'conversational' or 'spontaneous', which presents a significant challenge to ASR systems; moreover, meetings tend to involve many OOV terms.  ... 
doi:10.1145/1878101.1878107 fatcat:6uy6sq2l2jewznfu6d7ayqcfs4

The effect of language models on phonetic decoding for spoken term detection

Roy Wallace, Brendan Baker, Robbie Vogt, Sridha Sridharan
2009 Proceedings of the third workshop on Searching spontaneous conversational speech - SSCS '09  
This work challenges the assumption that improved speech recognition accuracy implies better indexing for STD.  ...  Spoken term detection (STD) popularly involves performing word or sub-word level speech recognition and indexing the result.  ...  This uni-gram word language model is trained from the same SWB data as the phonotactic language models plus 285 hrs of transcripts from the Fisher conversational telephone speech corpus.  ... 
doi:10.1145/1631127.1631132 fatcat:n4grdwnjr5fkjgokvlzsemthtm

Towards methods for efficient access to spoken content in the ami corpus

Gareth J. F. Jones, Maria Eskevich, Ágnes Gyarmati
2010 Proceedings of the 2010 international workshop on Searching spontaneous conversational speech - SSCS '10  
While most existing work on speech search focused on clearly defined document units, in this paper we describe our initial investigation into search of meeting content using the AMI meeting collection.  ...  A known-item search task is then performed using presentation slides from the meetings as search queries to locate relevant sections of the meetings.  ...  Search of meetings is an interesting task for spontaneous speech search since it incorporates all the issues highlighted above.  ... 
doi:10.1145/1878101.1878108 fatcat:awhbg7ohlnejxmxpbctqrzyewi

Novel methods for query selection and query combination in query-by-example spoken term detection

Javier Tejedor, Igor Szöke, Michal Fapso
2010 Proceedings of the 2010 international workshop on Searching spontaneous conversational speech - SSCS '10  
Query-by-example (QbE) spoken term detection (STD) is necessary for low-resource scenarios where training material is hardly available and word-based speech recognition systems cannot be employed.  ...  The second presents a novel feature level example combination to construct a more robust query used during the search.  ...  Feature extraction The system is trained and tested on telephone conversational speech (8kHz data). Fig. 3 presents the feature extraction used.  ... 
doi:10.1145/1878101.1878106 fatcat:l26dxwtaqbhffccsjztt3wwbpq
« Previous Showing results 1 — 15 out of 48,547 results