Search
Search Funnelback University
1 -
40 of
40
search results for `hours of broadcast news` |u:mi.eng.cam.ac.uk
Fully-matching results
-
MultiMedia Document Retrieval (1997-2000)
mi.eng.cam.ac.uk/research/projects/Multimedia_Document_Retrieval/7 Oct 2001: To illustrate the effectiveness of the above by extending the existing VMR demonstration system (produced within the VMR project GR/H87629) to support the interactive retrieval of broadcast news.. ... Using a fast recogniser we transcribed the 500 hours -
The CUHTK-Entropic 10xRT Broadcast News Transcription System
mi.eng.cam.ac.uk/reports/full_html/odell_darpa99.html/1 Mar 2000: ABSTRACT. This paper describes the development of the CUHTK-Entropic 10xRT Broadcast News Transcription System. ... This involved transcribing 100 hours of broadcast news material and this was infeasible with such a computationally expensive system. -
MultiMedia Document Retrieval (1997-2000) - Progress
mi.eng.cam.ac.uk/research/projects/Multimedia_Document_Retrieval/progress.html7 Oct 2001: This consisted of two parts. Firstly the automatic transcription of 100 hours of broadcast news data and secondly the retrieval of documents relevant to 23 natural language queries. ... Using a fast recogniser we transcribed the 500 hours of TREC-8 -
INDICATOR VARIABLE DEPENDENT OUTPUT PROBABILITY MODELLING…
mi.eng.cam.ac.uk/~sjy/papers/tuyo01.pdf20 Feb 2018: The following recognition experiments were conducted onthe 1997 Broadcast News task [5] by rescoring tri-gram lat-tices. ... Different topologies of themodel were evaluated on the 1997 Broadcast News task andsome small improvements were obtained. -
Experiments in Broadcast News Transcription
mi.eng.cam.ac.uk/reports/full_html/woodland_icassp98.html/1 Mar 2000: This corpus will be referred to as BNtrain96. A further tranche of data of similar size was released in 1997 to form in total 72 hours of broadcast news training data. ... Gauvain J.L., Lamel L., Adda G.& Adda-Decker M. (1997). Transcription of -
Building Multiple Complementary Systems using Directed Decision Trees …
mi.eng.cam.ac.uk/~mjfg/breslin_INTER07.pdf11 Jan 2008: end. Figure 4: Calculating Decision Tree Divergence. 5. ResultsExperiments were performed on a Broadcast News Arabic task.Each system was trained using 101.8 hours of data and a PLPfrontend. ... Results are given on three test sets: bnat05 (5.72 -
Who Really Spoke When? Finding Speaker Turns and Identities in…
mi.eng.cam.ac.uk/reports/full_html/tranter_icassp06.html/9 Dec 2006: Several methods can be employed to try to ascertain the true identity of the speakers within a particular Broadcast News show. ... The training data used for this task consisted of the Hub-4 1996/7 broadcast news training data. -
The 1998 HTK Broadcast News Transcription System: Development and…
mi.eng.cam.ac.uk/reports/full_html/woodland_darpa99.html/2 Mar 2000: Significant progress in the accurate transcription of broadcast news data has been made over the last few years so that we are now at a point where such systems can be ... The 1997 system used N-gram language models trained on 132 million words of -
TRAINING AND ADAPTING MLP FEATURES FOR ARABIC SPEECH RECOGNITION
mi.eng.cam.ac.uk/research/projects/AGILE/publications/park_icassp09.pdf7 Oct 2009: For large systems trained on hundreds of hours of data,this can be a significant cost. ... All thesetest sets consist of both Broadcast News and Broadcast Conversationstyles of data. -
Explicitly Generating Complementary Systems for Large…
mi.eng.cam.ac.uk/~mjfg/breslin_INTER06.pdf22 Nov 2006: This algorithm is described in detail in the next section,followed by preliminary results on a Broadcast News Mandarinsystem, before conclusions are drawn. ... 3. Experimental ResultsExperiments were performed on a Broadcast News Mandarin task.The -
The 1997 HTK Broadcast News Transcription System
mi.eng.cam.ac.uk/reports/full_html/woodland_darpa98.html/1 Mar 2000: This corpus will be referred to as BNtrain96. A further tranche of data of similar size was released in 1997 to form in total 72 hours of broadcast news training data. ... Gauvain J.L., Lamel L., Adda G.& Adda-Decker M. (1997). Transcription of -
DISCRIMINATIVE LANGUAGE MODEL ADAPTATION FORMANDARIN BROADCAST SPEECH …
mi.eng.cam.ac.uk/research/projects/AGILE/publications/liu-asru07.pdf26 Mar 2008: A total of 942 hours of broad-cast news (BN) and broadcast conversation (BC) speech audio datawere used for acoustic model training. ... Three Mandarin ASR evaluation sets are used:. • bnmdev06: 14 shows, 3.4 hours of BN data broadcast be-tween -
Abstract for johnson_trec8
mi.eng.cam.ac.uk/reports/abstracts/johnson_trec8.html27 Jul 2020: The 500 hours of broadcast news audio was filtered using an automatic scheme for detecting commercials, and then transcribed using a 2-pass HTK speech recogniser which ran at 13 times ... The final system gave an Average Precision of 55.29% on our -
IMPROVING BROADCAST NEWS TRANSCRIPTION BY LIGHTLY…
mi.eng.cam.ac.uk/reports/svr-ftp/chan_icassp2004.pdf27 May 2004: ABSTRACT. In this paper, we present our experiments on lightly superviseddiscriminative training with large amounts of broadcast news datafor which only closed caption transcriptions are available (TDTdata). ... Rich Tran-scription Workshop, 2003. [4] D. -
Abstract for johnson_icassp99
mi.eng.cam.ac.uk/reports/abstracts/johnson_icassp99.html27 Jul 2020: Woodland. March 1999. This paper describes the spoken document retrieval system that we have been developing and assesses its performance using automatic transcriptions of about 50 hours of broadcast news data. ... The recognition engine is based on the -
The Cambridge University Spoken Document Retrieval System
mi.eng.cam.ac.uk/reports/full_html/johnson_icassp99.html/8 Mar 2000: its performance using automatic transcriptions of about 50 hours of broadcast news data. ... For the 1998 TREC-7 SDR task about 100 hours of broadcast news had to be transcribed but the full 1997 Hub4 training set was available for HMM estimation. -
EXPERIMENTS IN BROADCAST NEWS TRANSCRIPTION P.C. Woodland, T. Hain,…
mi.eng.cam.ac.uk/reports/svr-ftp/woodland_icassp98.pdf10 Apr 2000: This corpus will be referred to as BN-train96. A further tranche of data of similar size was released in1997 to form in total 72 hours of broadcast news training ... 1997).Transcription of Broadcast News. Proc. Eurospeech’97, pp.907-910, Rhodes. [4] -
The 1998 HTK Broadcast News Transcription System:Development and…
mi.eng.cam.ac.uk/reports/svr-ftp/woodland_darpa99.pdf8 Mar 2000: The1997 system used N-gram language models trained on 132million words of broadcast news texts, the LDC-distributed1995 newswire texts, and the transcriptions from BNtrain97(LMtrain97). ... Furthermore we pro-cessed additional transcriptions of broadcast -
The Cambridge University Multimedia Document Retrieval Demo System
mi.eng.cam.ac.uk/reports/full_html/tuerk_riao00demo.html/14 Aug 2000: The system is trained on about 150 hours of acoustic training data and 260 million words of broadcast news and newspaper transcriptions. ... The system gives a word error rate of 15.9% on the 1998 Hub4 broadcast news evaluation data. -
DISCRIMINATIVE LANGUAGE MODEL ADAPTATION FORMANDARIN BROADCAST SPEECH …
mi.eng.cam.ac.uk/~mjfg/liu-asru07.pdf11 Jan 2008: A total of 942 hours of broad-cast news (BN) and broadcast conversation (BC) speech audio datawere used for acoustic model training. ... Three Mandarin ASR evaluation sets are used:. • bnmdev06: 14 shows, 3.4 hours of BN data broadcast be-tween -
Cambridge STT Overview P.C. Woodland, H.Y. Chan, G. Evermann, ...
mi.eng.cam.ac.uk/research/projects/EARS/pubs/woodland_earsfeb04.pdf23 Mar 2004: Woodland et al.: Cambridge STT Overview. Outline. • Broadcast News. • Lightly supervised discriminative training. • ... Lightly supervised discriminative training on TDT data. • Improve the English Broadcast News system by adding large amounts of -
Effects of Out of Vocabulary Words in Spoken Document Retrieval
mi.eng.cam.ac.uk/reports/full_html/woodland_sigir00.html/14 Aug 2000: The TREC-8 audio contains 500 hours of US broadcast news data that was recorded between February and June 1998. ... Transcription used a simplified version of the HTK broadcast news system which corresponds to the first-pass'' recognition system -
Recent Progress in Large Vocabulary ContinuousSpeech Recognition: An…
mi.eng.cam.ac.uk/~mjfg/icassp06_tutorial.pdf22 Feb 2007: Broadcast News (BN)– Single audio stream with many talkers, styles, noise conditions, bandwiths– Much of it prepared speech from anchor speakers but some conversational– Need to segment for normalisation/adaptation– For ... English: 200h of -
Annotating large lattices with the exact word error Rogier ...
mi.eng.cam.ac.uk/~mjfg/Kernel/van_dalen-2015-exact_error.pdf16 Jun 2015: The training setwas 34 hours of randomly selected shows from the ’96 releaseof the Hub4 broadcast news, LDC97S44 [17]. ... Pallett, “1996English broadcast news speech (HUB4), LDC97S44,” Philadel-phia, 1997. [18] M. -
SPOKEN DOCUMENT RETRIEVAL FOR TREC-8 AT CAMBRIDGE UNIVERSITY S.E. ...
mi.eng.cam.ac.uk/reports/svr-ftp/johnson_trec8.pdf10 Apr 2000: The 500hours of broadcast news audio was filtered using an automaticscheme for detecting commercials, and then transcribed using a2-pass HTK speech recogniser which ran at 13 times real time.The ... The HMMs were trained using 146 hours of broadcast news -
This page has been left blank. SPOKEN DOCUMENT RETRIEVAL ...
mi.eng.cam.ac.uk/reports/svr-ftp/johnson_trec9.pdf29 Jul 2001: of American news broadcast between. ... DARPA Broadcast News Transcriptionand Understanding Workshop, pp. 133-137, 1998. [12] S.E. -
SPOKEN DOCUMENT RETRIEVAL FOR TREC-7 AT CAMBRIDGE UNIVERSITY S.E. ...
mi.eng.cam.ac.uk/reports/svr-ftp/johnson_trec7.pdf8 Mar 2000: TREC-7 1. 2. THE HTK BROADCAST NEWS TRANSCRIPTIONSYSTEM. The input data is presented to our HTK transcription system as com-plete episodes of broadcast news shows and these are first ... The HMMs for TREC-7 used HMMs trained on 70 hours of acous-tic data -
Article Submitted to Computer Speech and Language Automatic…
mi.eng.cam.ac.uk/reports/svr-ftp/auto-pdf/kim_csl04.pdf9 Aug 2005: broadcast news and conversational speech over thetelephone, explicit indications of capitalised words are not given. ... Broadcast News provides a good test-bed for speech recognition, because it requiressystems to handle a wide range of speakers, a -
General Query Expansion Techniques for Spoken Document Retrieval
mi.eng.cam.ac.uk/reports/full_html/jourlin_esca99.html/2 Mar 2000: The input data is presented to the system as complete episodes of broadcast news shows and these are first converted to a set of segments for further processing [9]. ... The HMMs used in TREC-7 were trained on 70 hours of acoustic data and the language -
Structured Deep Neural Networks for Speech Recognition
mi.eng.cam.ac.uk/~mjfg/thesis_cw564.pdf12 Jul 2018: Stimulated DNNswere trained using the KL activation regularisation. 135. 7.6 Broadcast News: Summary of training and evaluation sets, includingtotal hours, number of utterances and average utterance duration. ... 139. 7.9 Broadcast News: Recognition -
Spoken Document Retrieval for TREC-8 at Cambridge University
mi.eng.cam.ac.uk/reports/full_html/johnson_trec8.html/30 Mar 2000: The 500 hours of broadcast news audio was filtered using an automatic scheme for detecting commercials, and then transcribed using a 2-pass HTK speech recogniser which ran at 13 times ... The HMMs were trained using 146 hours of broadcast news audio -
Generation and Combination ofComplementary Systems for Automatic…
mi.eng.cam.ac.uk/~mjfg/thesis_cb404.pdf9 Jul 2008: 1. CHAPTER 1. INTRODUCTION 2. character error rate (CER) on broadcast news Mandarin [48], which is significantly worse thanthe performance of human transcription on spontaneous speech [97]. ... Recent LVCSR projects include the AGILE/GALE project1 for -
Uncertainty Decoding forNoise Robust Speech Recognition Hank Liao…
mi.eng.cam.ac.uk/~mjfg/thesis_hl251.pdf17 Sep 2008: 1208.2.4 Combined Systems. 122. 8.3 Summary. 124. 9 Experimental Results on Recorded Noisy Speech 1259.1 Broadcast News Transcription. ... as Broadcast News and Toshiba Research Europe’s internal collection of in-carspeech data. -
PhD Thesis
mi.eng.cam.ac.uk/~mjfg/thesis_kcs23.pdf16 Nov 2007: BN Broadcast News. BW Baum Welch. CDHMM Continuous Density Hidden Markov Model. ... more recent Conversational Telephone Speech (CTS) and Broadcast News (BN) data sets. -
Named Entity Recognition from Speechand Its Use in the ...
mi.eng.cam.ac.uk/reports/svr-ftp/auto-pdf/kim_thesis.pdf9 Aug 2005: for the 1998 NIST Hub-4 Information Extraction (Named Entity) Broadcast News Benchmark. ... Focus-. ing on the 1999 DARPA Broadcast News Workshop proceedings, which contain the results of the. -
Statistical Machine Translationand Automatic Speech Recognitionunder…
mi.eng.cam.ac.uk/~wjb31/ppubs/LMathiasDissDec07.pdf16 Feb 2008: hours of speech data, for example from broadcast news television feeds. ... Although, it is easy to obtain several hours. of speech data, obtaining the corresponding transcripts is a time consuming and. -
Audio Indexing and Retrieval of Complete Broadcast News Shows
mi.eng.cam.ac.uk/reports/full_html/johnson_riao00.html/19 Apr 2000: This is particularly important for the case of broadcast news since the density of important up-to-date information is generally high, but topic changes occur frequently and information on a ... The experiments reported in this paper use the framework of -
Spoken Document Retrieval for TREC-7 at Cambridge University
mi.eng.cam.ac.uk/reports/full_html/johnson_trec7.html/30 Mar 2000: The input data is presented to our HTK transcription system as complete episodes of broadcast news shows and these are first converted to a set of segments for further processing. ... The HMMs for TREC-7 used HMMs trained on 70 hours of acoustic data and -
Spoken Document Retrieval for TREC-9 at Cambridge University
mi.eng.cam.ac.uk/reports/full_html/johnson_trec9.html/23 Feb 2002: D. Abberley, S. Renals, G. Cook & T. Robinson. Retrieval of Broadcast News Documents with the THISL System. ... Jourlin, G.L.Moore, K. Spärck Jones & P.C. Woodland. Audio Indexing and Retrieval of Complete Broadcast News Shows. -
THE CAMBRIDGE UNIVERSITY SPOKEN DOCUMENT RETRIEVAL SYSTEM S.E.…
mi.eng.cam.ac.uk/reports/svr-ftp/johnson_icassp99.pdf8 Mar 2000: hours of broadcast news data. ... of broadcast news texts, the LDC-distributed 1995newswire texts, and the transcriptions of the acoustic training data.
Search history
Recently clicked results
Recently clicked results
Your click history is empty.
Recent searches
- `Watson Smith` |u:www.reporter.admin.cam.ac.uk (105) · moments ago
Recent searches
Your search history is empty.