Search

Search Funnelback University

Search powered by Funnelback
1 - 50 of 58 search results for news |u:mi.eng.cam.ac.uk
  1. Fully-matching results

  2. The Cambridge University Multimedia Document Retrieval Demo System

    mi.eng.cam.ac.uk/reports/full_html/tuerk_sigir00demo.html/
    14 Aug 2000: This system gives a word error rate of 15.9% on the 1998 Hub4 broadcast news evaluation data. ... J. J. Odell, P. C. Woodland, and T. Hain. The CUHTK-Entropic 10xRT Broadcast News Transcription System.
  3. The Cambridge University Multimedia Document Retrieval Demo System

    mi.eng.cam.ac.uk/reports/full_html/tuerk_riao00demo.html/
    14 Aug 2000: The system gives a word error rate of 15.9% on the 1998 Hub4 broadcast news evaluation data. ... Also a windowing system is being developed that allows automatic content dependent segmentation of news broadcasts.
  4. Improving Retrieval on Imperfect Speech Transcriptions

    mi.eng.cam.ac.uk/reports/full_html/jourlin_sigir99.html/
    8 Mar 2000: This paper presents the results from adding several forms of query expansion to our retrieval system running on several sets of transcriptions of broadcast news from the 1997 TREC-7 spoken ... A parallel corpus of 18628 documents from manually
  5. The CUHTK-Entropic 10xRT Broadcast News Transcription System

    mi.eng.cam.ac.uk/reports/full_html/odell_darpa99.html/
    1 Mar 2000: The CUHTK-Entropic 10xRT Broadcast News Transcription System. J.J. Odell , P.C. ... ABSTRACT. This paper describes the development of the CUHTK-Entropic 10xRT Broadcast News Transcription System.
  6. Segment Generation and Clustering in the HTK Broadcast News

    mi.eng.cam.ac.uk/reports/full_html/hain_darpa98.html/
    1 Mar 2000: 133-137 (Lansdowne, VA, Feb. 1998). SEGMENT GENERATION AND CLUSTERING IN THE HTK BROADCAST NEWS TRANSCRIPTION SYSTEM. ... The distribution of broadcast news data suitable for GMM training can be seen in Table 1.
  7. Experiments in Broadcast News Transcription

    mi.eng.cam.ac.uk/reports/full_html/woodland_icassp98.html/
    1 Mar 2000: Experiments in Broadcast News Transcription. P.C. Woodland, T. Hain, S.E. Johnson, T. ... FX. all other speech (e.g. spontaneous non-native). Table 1: Broadcast news focus conditions.
  8. 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
  9. Speaker Clustering Using Direct Maximisation of the MLLR-Adapted…

    mi.eng.cam.ac.uk/reports/full_html/johnson_icslp98.html/
    8 Mar 2000: Use of these new clustering schemes has been shown to improve word error rates on the Hub4 1997 broadcast news evaluation data. ... M A Siegler, U Jain, B Raj & R M Stern. Automatic Segmentation, Classification and Clustering of Broadcast News [pp.
  10. The 1998 HTK Broadcast News Transcription System: Development and…

    mi.eng.cam.ac.uk/reports/full_html/woodland_darpa99.html/
    2 Mar 2000: FX. all other speech (e.g. spontaneous non-native). Table 1: Broadcast news focus conditions. ... Proc. 1998 DARPA Broadcast News Transcription and Understanding Workshop, pp. 41-48, Lansdowne.
  11. Global Optimisation of Neural Network Models Via Sequential…

    mi.eng.cam.ac.uk/reports/full_html/freitas_icslp98.html/
    8 Mar 2000: We show how the new algorithms can outperform extended Kalman filter (EKF) training. ... weights (eqn 1), evaluate the new importance ratios (eqns 3,4,5) and resample if necessary.
  12. The Cambridge University Spoken Document Retrieval System

    mi.eng.cam.ac.uk/reports/full_html/johnson_icassp99.html/
    8 Mar 2000: Unfortunately the Porter algorithm suffers from some problems for example mapping news'' to new'' but not equating government'' and governmental''. ... Furthermore new language models were estimated from an extended set of broadcast news transcripts and
  13. A Method for Direct Audio Search with Applications to Indexing and…

    mi.eng.cam.ac.uk/reports/full_html/johnson_icassp00.html/
    16 Jun 2000: Method. CPUs. non-news. news. window, 5@1s. 3514. 5459s=60.8%. 110s=0.5%. window, 10@2s. ... windows. (s). non-news. news. 5s long, 1s shift. 0. 60.8%. 0.49%.
  14. Audio Indexing and Retrieval of Complete Broadcast News Shows

    mi.eng.cam.ac.uk/reports/full_html/johnson_riao00.html/
    19 Apr 2000: 1163-1177 (Paris,France, April 2000). Audio Indexing and Retrieval of Complete Broadcast News Shows. ... Dharanipragada, S., Franz, M., and Roukos, S. (1999). Audio-Indexing for Broadcast News.
  15. Who Spoke When? - Automatic Segmentation and Clustering for…

    mi.eng.cam.ac.uk/reports/full_html/johnson_eurospeech99.html/
    2 Mar 2000: The 1996 Hub-4 Broadcast News Transcription development data was used for all the experiments reported in this paper. ... A new frame-based efficiency is therefore defined. The previous formulae remain the same but the definitions are altered to:.
  16. The Cambridge University Multimedia Document RetrievalDemo System…

    mi.eng.cam.ac.uk/reports/svr-ftp/tuerk_sigir00demo.pdf
    3 May 2000: C. Woodland, and T. Hain. The CUHTK-Entropic 10xRT Broadcast News Transcription System. ... InProc. DARPA Broadcast News Workshop, pages 271–275,Herndon, VA, 1999. [3] A.
  17. The Cambridge University Multimedia DocumentRetrievalDemoSystem…

    mi.eng.cam.ac.uk/reports/svr-ftp/tuerk_riao00demo.pdf
    20 Apr 2000: 2 Systemdescription. 2.1 Overview. The CU-MDR demosystemdownloadsthe audio track of news broadcastsfrom the Internetoncea day and addsthem to its archive. ... Also a windowing systemis beingdevelopedthatallows automaticcontentdependentsegmentationof news
  18. General Query Expansion Techniques for Spoken Document Retrieval

    mi.eng.cam.ac.uk/reports/full_html/jourlin_esca99.html/
    2 Mar 2000: The new document universe is defined with a set of query atomic terms. ... An example at this stage would look like :. Query. : "Trains in New York".
  19. Class-based language model adaptation using mixtures of word-class…

    mi.eng.cam.ac.uk/reports/full_html/moore_icslp00.html/
    2 Nov 2000: S. Sekine, Automatic Sublanguage Identification for a New Text''; Second Annual Workshop on Very Large Corpora, Kyoto. ... Whittaker and S.J. Young, The 1997 HTK Broadcast News Transcription System''; DARPA Broadcast News 1998 Transcription and
  20. The 1997 HTK Broadcast News Transcription System

    mi.eng.cam.ac.uk/reports/full_html/woodland_darpa98.html/
    1 Mar 2000: 41-48 (Lansdowne,VA, Feb. 1998). The 1997 HTK BROADCAST NEWS TRANSCRIPTION SYSTEM. ... using the broadcast news training texts, the acoustic training data and 1995 Marketplace transcriptions.
  21. POSTERIOR PROBABILITY DECODING, CONFIDENCE ESTIMATION AND SYSTEM…

    mi.eng.cam.ac.uk/reports/full_html/evermann_stw00.html/
    5 Oct 2000: In this paper, a new approach that is a generalisation of the ROVER technique is presented. ... Finally the new system combination scheme used in the CU-HTK evaluation system is presented in section 6..
  22. The CUHTK-Entropic 10xRT Broadcast News Transcription SystemJ.J.…

    mi.eng.cam.ac.uk/reports/svr-ftp/odell_darpa99.pdf
    8 Mar 2000: The CUHTK-Entropic 10xRT Broadcast News Transcription SystemJ.J. Odell. , P.C. Woodland. & ... ror rate for systems running in less than 10xRT in the 1998DARPA broadcast news evaluation.
  23. LARGE SCALE MMIE TRAINING FOR CONVERSATIONAL TELEPHONE SPEECH…

    mi.eng.cam.ac.uk/reports/full_html/woodland_stw00.html/
    5 Oct 2000: The system uses word-based N-gram LMs estimated from an interpolation of Hub5 acoustic training transcriptions and Broadcast News texts.
  24. Segment Generation and Clustering in the HTKBroadcast News

    mi.eng.cam.ac.uk/reports/svr-ftp/hain_darpa98.pdf
    8 Mar 2000: The distribution of broadcast news data suitable for GMM trainingcan be seen in Table 1. ... Young S.J. (1997) TheDevelopment of the 1996 Broadcast News Transcription Sys-tem.
  25. EXPERIMENTS IN BROADCAST NEWS TRANSCRIPTION P.C. Woodland, T. Hain,…

    mi.eng.cam.ac.uk/reports/svr-ftp/woodland_icassp98.pdf
    10 Apr 2000: EXPERIMENTS IN BROADCAST NEWS TRANSCRIPTION. P.C. Woodland, T. Hain, S.E. Johnson, T.R. ... Young S.J. (1997) TheDevelopment of the 1996 Broadcast News Transcription Sys-tem.
  26. THE CU-HTK MARCH 2000 HUB5E TRANSCRIPTION SYSTEM

    mi.eng.cam.ac.uk/reports/full_html/hain_stw00.html/
    12 Oct 2000: This gave a new word list with 54,537 words where most of the pronunciations were already available in our broadcast news (Hub4) dictionary. ... P.C. Woodland, M.J.F. Gales, D. Pye & S.J. Young (1997). Broadcast News Transcription Using HTK.
  27. The 1998 HTK Broadcast News Transcription System:Development and…

    mi.eng.cam.ac.uk/reports/svr-ftp/woodland_darpa99.pdf
    8 Mar 2000: The 1998 HTK Broadcast News Transcription System:Development and Results. P.C. Woodland, T. ... Proc. 1998 DARPA Broad-cast News Transcription and Understanding Workshop, pp.133-137. 5.
  28. Effects of Out of Vocabulary Words in Spoken Document ...

    mi.eng.cam.ac.uk/reports/svr-ftp/woodland_sigir00.pdf
    10 May 2000: Thespeech recogniser uses cross-word context-dependent hid-den Markov models and a 4-gram language model trainedon broadcast news data. ... Table 3 gives the performance for BRF when adding3 new terms drawn from the top 10 documents.
  29. Spoken Document Retrieval for TREC-7 at Cambridge University

    mi.eng.cam.ac.uk/reports/full_html/johnson_trec7.html/
    30 Mar 2000: This was useful, but we went further and developed a new pair of stoplists specifically for the broadcast news data. ... to be made equivalent. A stemming exceptions list was also made to compensate for known problems with the Porter stemmer, such as
  30. Spoken Document Retrieval for TREC-8 at Cambridge University

    mi.eng.cam.ac.uk/reports/full_html/johnson_trec8.html/
    30 Mar 2000: news programs provided along with a list of manually-generated story (or document) boundaries. ... 1996 to Jan. 1998 (from the Primary Source Media Broadcast News collection) and Jan.
  31. SPEAKER CLUSTERING USING DIRECT MAXIMISATION OFTHE MLLR-ADAPTED…

    mi.eng.cam.ac.uk/reports/svr-ftp/johnson_icslp98.pdf
    10 Apr 2000: Use of these new clusteringschemes has been shown to improve word error rates on theHub4 1997 broadcast news evaluation data. ... AutomaticSegmentation, Classification and Clustering of Broad-cast News. Proc. 1997 DARPA Speech RecognitionWorkshop, pp.
  32. THE DEVELOPMENT OF THE1996 HTK BROADCAST NEWS TRANSCRIPTION SYSTEM ...

    mi.eng.cam.ac.uk/reports/svr-ftp/woodland_darpa97.pdf
    8 Mar 2000: THE DEVELOPMENT OF THE1996 HTK BROADCAST NEWS TRANSCRIPTION SYSTEM. P.C. Woodland, M.J.F. ... 5. CONCLUSIONThis paper has described our initial efforts to develop systemsfor broadcast news transcription.
  33. GLOBAL OPTIMISATION OF NEURAL NETWORK MODELSVIA SEQUENTIAL…

    mi.eng.cam.ac.uk/reports/svr-ftp/freitas_icslp98.pdf
    10 Apr 2000: Thus we can draw initial weights and importance ratiosfrom the prior (eqn 6) and for each sampling stage, pre-dict the new weights (eqn 1), evaluate the new importanceratios (eqns ... By repeating this S times, the newresampled indices are determined,
  34. A METHOD FOR DIRECT AUDIO SEARCH WITH APPLICATIONS TOINDEXING ...

    mi.eng.cam.ac.uk/reports/svr-ftp/johnson_icassp00.pdf
    19 Apr 2000: 75 73.1% 0.79%. Table 5: News and non-news rejection when smoothing be-tween repeated audio. ... Speeding up the Search. Reducing the Size of Cue-Audio. Expt. 2 -- Finding Non-News Events.
  35. A Statistical Consistency Check for the SpaceCarving Algorithm. A. ...

    mi.eng.cam.ac.uk/reports/svr-ftp/broadhurst_cipolla_bmvc2000.pdf
    25 Oct 2000: A new consistencyfunction is proposedthat usesa statisticalcomparisoninsteadof the voxelcentroidsamplingthat wasinitially proposed.This is importantwhenthereis moredetailin theimagesthancanbestoredin a voxel representation.Thenew functionis
  36. THE 1998 HTK SYSTEM FOR TRANSCRIPTION OFCONVERSATIONAL TELEPHONE…

    mi.eng.cam.ac.uk/reports/svr-ftp/hain_icassp99.pdf
    27 Sep 2000: Subsequently we present thespeaker adaptation tests using the new front-end, followed by de-tails about language models used. ... Young S.J. (1998) The 1997HTK Broadcast News Transcription System.Proc. DARPABroadcast News Transcription and Understanding
  37. THE CAMBRIDGE UNIVERSITY SPOKEN DOCUMENT RETRIEVAL SYSTEM S.E.…

    mi.eng.cam.ac.uk/reports/svr-ftp/johnson_icassp99.pdf
    8 Mar 2000: Unfortunately thePorter algorithm suffers from some problems for example mapping“news” to “new” but not equating “government” and “governmen-tal”. ... We therefore used a new set of HTKtriphone HMMs in an HTK-2 recognition setup.
  38. Class-based language model adaptation using mixtures ofword-class…

    mi.eng.cam.ac.uk/reports/svr-ftp/moore_icslp00.pdf
    2 Nov 2000: 2 Finding the TopicsA mixture of broadcast news and newswire text was usedas training data for the topic model, with 144 million wordsof Broadcast News text and 25 million words of ... Whittaker and S.J. Young, “The 1997 HTKBroadcast News Transcription
  39. WHO SPOKE WHEN? - AUTOMATIC SEGMENTATION ANDCLUSTERING FOR…

    mi.eng.cam.ac.uk/reports/svr-ftp/johnson_eurospeech99.pdf
    10 Apr 2000: These methods were shown tobe an important part of our overall recognition systemfor broadcast news. ... The 1996 Hub-4 Broadcast News Transcription develop-ment data was used for all the experiments reported inthis paper.
  40. LARGE VOCABULARY DECODING AND CONFIDENCE ESTIMATIONUSING WORD…

    mi.eng.cam.ac.uk/reports/svr-ftp/evermann_icassp00.pdf
    5 May 2000: It is also interesting tonote that the improvement is consistent over the varioustypes of data found in broadcast news. ... Johnson, T.R. Niesler,A. Tuerk, E.W.D. Whittaker, and S.J. Young. The1997 HTK Broadcast News Transcription System.
  41. The 1997 HTK Broadcast News Transcription SystemP.C. Woodland, T. ...

    mi.eng.cam.ac.uk/reports/svr-ftp/woodland_darpa98.pdf
    8 Mar 2000: The 1997 HTK Broadcast News Transcription SystemP.C. Woodland, T. Hain, S.E. ... news development test data andjust 15.8% on the 1997 evaluation test set.
  42. The Applications of Uncalibrated Occlusion Junctions A.…

    mi.eng.cam.ac.uk/reports/svr-ftp/broadhurst_cipolla_bmvc1999.pdf
    25 Oct 2000: This work is extended,sothatjunctionpointscanbe representedfor arbitraryprojective cameras,andusingthis representation,a new constraint,the trifocal junction point constraint will be presented.This constraintis usedin an algorithmto ... 10). A
  43. Reconstruction and Motion Estimation fromApparent Contours under…

    mi.eng.cam.ac.uk/reports/svr-ftp/wong_bmvc99.pdf
    30 Jan 2000: Nevertheless,thepresenceof local minimaandtheinsensitivity ofthecostfunctionto variationsin make thesearchvery difficult.point, and a new ellipse was fitted back to the disturbedpoints.
  44. GENERAL QUERY EXPANSION TECHNIQUESFOR SPOKEN DOCUMENT RETRIEVAL…

    mi.eng.cam.ac.uk/reports/svr-ftp/jourlin_esca99.pdf
    10 Apr 2000: different sets ofautomatically transcribed broadcast news texts with aWER varying from24.8% to 66.2%. ... This infor-mation can be used to reweight existing query wordsor add new ones.
  45. ITERATIVE UNSUPERVISED ADAPTATION USING MAXIMUMLIKELIHOOD LINEAR…

    mi.eng.cam.ac.uk/reports/svr-ftp/woodland_icslp96.pdf
    17 Oct 2000: acoustic informationand can be used for rescoring with new acoustic models, or for theapplication of new language models. ... Each of20 speakers read 15 sentences from one news article. The test wasdefined so that data for each speaker (or session) could
  46. POSTERIOR PROBABILITY DECODING, CONFIDENCEESTIMATION AND SYSTEM…

    mi.eng.cam.ac.uk/reports/svr-ftp/evermann_stw00.pdf
    5 Oct 2000: A new system combination techniqueis presented that uses these confidence scores and the confusion net-works and performs better than the well-known ROVER technique. ... In this paper, a new approach thatis a generalisation of the ROVER technique is
  47. FRAME DISCRIMIN ATION TRAINING OF HMMS FORLARGE VOCABULAR Y ...

    mi.eng.cam.ac.uk/reports/svr-ftp/povey_tr332.pdf
    6 Sep 2000: 7:9h , } h , -} h are respectively the numeratorand denominatormixtureoccupanciesand the old and new mixture weights. ... Of course,this is not pos-sible becauseit cannotbe known exactly what the new occupancieswill be.
  48. Audio Indexing and Retrieval of Complete Broadcast News Shows ...

    mi.eng.cam.ac.uk/reports/svr-ftp/johnson_riao00.pdf
    10 Apr 2000: Audio Indexing and Retrieval of Complete Broadcast News Shows. S.E. Johnson†, P. ... InProc. 1999 DARPA Broadcast News Workshop, pages 271–275, Hern-don, VA. Robertson, S.
  49. THE CU-HTK MARCH 2000 HUB5E TRANSCRIPTION SYSTEMT. Hain, P.C. ...

    mi.eng.cam.ac.uk/reports/svr-ftp/hain_stw00.pdf
    5 Oct 2000: 50,000 words occurring in the 204 millionwords of broadcast news (BN) LM training data. ... 18. P.C. Woodland, M.J.F. Gales, D. Pye & S.J. Young (1997).Broadcast News Transcription Using HTK.Proc.
  50. Recovery of Circular Motion from Profiles of Surfaces Paulo ...

    mi.eng.cam.ac.uk/reports/svr-ftp/mendonca_iccv99ws.pdf
    1 Jun 2000: 6 Conclusions and Future Work. This paper introduces a new method of motion estimation by using profiles of a rotat-ing object. ... 5. H. S. M. Coxeter. Introduction to Geometry. John Wiley and Sons, New York, secondedition, 1969.
  51. SPOKEN DOCUMENT RETRIEVAL FOR TREC-8 AT CAMBRIDGE UNIVERSITY S.E. ...

    mi.eng.cam.ac.uk/reports/svr-ftp/johnson_trec8.pdf
    10 Apr 2000: 1996 to Jan. 1998 (from the Primary Source MediaBroadcast News collection) and Jan. ... query.3. Perform BRF onIt and add the new terms to the query.

Refine your results

Related searches for news |u:mi.eng.cam.ac.uk

Search history

Recently clicked results

Recently clicked results

Your click history is empty.

Recent searches

Your search history is empty.