Search
Search Funnelback University
- Refined by:
- Date: 2005
1 -
50 of
190
search results for TALK:PC53 20 |u:mi.eng.cam.ac.uk
where 0
match all words and 190
match some words.
Results that match 1 of 2 words
-
Two-way Cluster Voting to Improve Speaker Diarisation Performance
mi.eng.cam.ac.uk/reports/full_html/tranter_icassp05.html/31 Mar 2005: 20.78. 19.15. 32.06. 37.18. 10.74. 24.52. † St. 128mix. 27.66. 20.78. ... 19.86. 11.78. 20.37. 22.77. 12.28. 18.10. † St. 128mix. 26.34. 18.21. -
Recent Improvementsin the CUED CTS SU System M. Tomalin ...
mi.eng.cam.ac.uk/research/projects/EARS/pubs/tomalin_mar05mde.pdf28 Apr 2005: 20.4 %: before co-ordinating conjunctionEx: that’s unbiased honesty () but then again. ... 20.4 %: before potential Discourse MarkerEx: on the other hand well by telling the truth. -
The Development of the Cambridge UniversityRT-04 Diarisation System…
mi.eng.cam.ac.uk/research/projects/EARS/pubs/tranter_rt04.pdf15 Feb 2005: sorting didev03 eval03 sttdev04 dev04f2 devallnone 18.0 15.9 21.2 26.9 20.3time 17.5 16.7 21.5 25.7 20.2spkr-start 17.5 17.9 ... 18.5. 19. 19.5. 20. mea. n D. ER. on. all. 24 d. -
The Development of the Cambridge University RT-04 Diarisation System
mi.eng.cam.ac.uk/reports/full_html/tranter_rt04.html/10 Jan 2005: 22.4. 20.2. PLP. Y. -. Y. 25.4. 19.3. 27.9. 24.1. 24.3. ... 16.9. 17.8. 24.6. 24.0. 20.9 @ 2521. 10. 10. 19.9. 21.3. -
The Cambridge University March 2005 Speaker Diarisation System
mi.eng.cam.ac.uk/reports/full_html/sinha_eurospeech05.html/22 Sep 2005: B D. models were formed by performing a single iteration of (mean-only) MAP adaptation, with τ =20 from the. ... IAC. DataSet. MS/FA/SPE/DER. Cluster Imp. Seg Imp. Base. Dev24. 1.2/1.1/17.9/20.17. 6.59@718. -
Progress in English Conversational TelephoneSpeech Transcription Khe…
mi.eng.cam.ac.uk/research/projects/EARS/pubs/sim_sttmar05.pdf12 Apr 2005: S6 15K (36) 32.1 24.3 28.3 24.3. S1 6K (28)8. 27.9 20.2 24.2 20.5S4 9K (36) 26.8 19.5 23.3 ... S4—. 826.7 19.6 23.3 20.0. 9K 26.3 18.9 22.7 19.4. S6 — 8 26.5 19.4 23.0 19.5. -
INVESTIGATION OF ACOUSTIC MODELING TECHNIQUES FOR LVCSR SYSTEMS X. ...
mi.eng.cam.ac.uk/reports/svr-ftp/liu_icassp2005.pdf19 May 2005: The finalerror rates were 20.5% on eval03 and 16.9% on dev04. ... P3aP3c 23.9 16.8 20.5 16.9. Table 1. CTS 10xRT system baseline performance. -
Advances in Structural Metadata for RT-04 atCUED M. Tomalin ...
mi.eng.cam.ac.uk/research/projects/EARS/pubs/tomalin_rt04.pdf15 Feb 2005: 0/56.8 29.2/15.7/56.2. PFMctsrt04 cl40-tg 33.1/20.3/63.9 33.3/18.7/62.6 30.8/19.7/61.9. ... Cambridge University RT-04 workshop: November 2004 20. Tomalin and Woodland: Advances in Structural Metadata for RT-04 at CUED. -
TWO-WAY CLUSTER VOTING TO IMPROVE SPEAKER DIARISATION PERFORMANCE S.…
mi.eng.cam.ac.uk/reports/svr-ftp/tranter_icassp05.pdf25 Mar 2005: 63 20.78 21.15 32.06 37.18 10.74 25.19EP 15mix 30.30 19.27 18.11 29.84 37.18 10.74 23.48. ... 34 18.21 12.06 30.78 33.62 8.99 21.54EP 16mix 27.01 19.66 10.45 26.88 33.62 8.99 20.33. -
Recent Improvements in the CUED DiarisationSystem Sue Tranter, Rohit…
mi.eng.cam.ac.uk/research/projects/EARS/pubs/tranter_mar05mde.pdf26 Mar 2005: 0 constant 19.1 17.1 (17.1)3 Concat. 0 BIC 18.9 20.3 (16.8)4 Concat. ... MDE Technical Meeting, March 2005 20. Tranter et al.: Recent Improvements in the CUED Diarisation System. -
Recent Developments at Cambridgein Broadcast News Transcription D.Y.…
mi.eng.cam.ac.uk/research/projects/EARS/pubs/kim_rt04.pdf15 Feb 2005: Cambridge UniversityEngineering Department. RT04 EARS workshop 20. Recent Developments at Cambridge in Broadcast News Transcription. ... 0.16 0.18 0.2 0.22 0.24 0.2612. 14. 16. 18. 20. 22. -
A NOVEL SELF-ORGANISING SPEECH PRODUCTIONSYSTEM USING…
mi.eng.cam.ac.uk/reports/svr-ftp/auto-pdf/blackburn_ICPhS95.pdf9 Aug 2005: 10 20 30 40 50 60 703.5. 3. 2.5. 2. 1.5. ... 0 20 40 60 80 1000. 8000. Speech frame index (10msec increments). -
��������� ��� ������� ��������� ��������� � ���� ���� ��� ...
mi.eng.cam.ac.uk/reports/svr-ftp/auto-pdf/saleh_tr205.pdf9 Aug 2005: 0. 0.1. 0.2Speech Segment. 0 5 10 15 20 25 302. ... 0 5 10 15 20 25 302. 1.5. 1. 0.5. 0. -
Face Recognition from Face Motion Manifolds using Robust…
mi.eng.cam.ac.uk/reports/svr-ftp/oa214_FPIV_2004_paper1.pdf8 Aug 2005: 5. 10. 15. 20. 25. 30. 35. 5 10 15 20 25 30 35. ... 20] W. S. Yambor. Analysis of PCA-based and fisherdiscriminant-based image recognition algorithms. -
THE USE OF RECURRENT NEURAL NETWORKSFOR CLASSIFICATION T. L. ...
mi.eng.cam.ac.uk/reports/svr-ftp/auto-pdf/burrows_nnsp94.pdf9 Aug 2005: 20 40 60 80 100 120 140 1601. 0.5. 0. 0.5. ... 1. Am. plit. ud. e. b) Hidden Node Outputs. 20 40 60 80 100 120 140 1601.5. -
The Cambridge University March 2005 Speaker Diarisation System R. ...
mi.eng.cam.ac.uk/reports/full_html/sinha_eurospeech05.html/paper.pdf22 Sep 2005: only) MAPadaptation, withτ = 20 from theB models to theD data. ... IAC DataSetMS/FA/SPE/DER Cluster Imp Seg ImpBase Dev24 1.2/1.1/17.9/20.176.59@718 4.36@2363Base Eval -
Rapid, Easy and ReliableCalibration for Freehand 3D Ultrasound P-W.…
mi.eng.cam.ac.uk/reports/svr-ftp/hsu_tr534.pdf16 Sep 2005: Ultrasound in Medicine & Biology, 20(9):923–936, 1994. [5] A. Fenster, D. ... IEEE Transactions on Medical Imaging, 17(5):681–693, 1998. [20] A. State, D. -
� ������� � ��� ����� ��������� ���������������������� ���…
mi.eng.cam.ac.uk/reports/svr-ftp/auto-pdf/niesler_tr258.pdf9 Aug 2005: number of times we would expect to see4. in7[. is. 0 2#4?5@7>.,8 : -)0;2#4D5N7>.,8/L 27>.,8 (20)Assuming words to occur independently within each -
CLASSIFICATION USING HIERARCHICALMIXTURES OF EXPERTS S.R.Waterhouse…
mi.eng.cam.ac.uk/reports/svr-ftp/auto-pdf/waterhouse_hme.pdf9 Aug 2005: In particular we use a 40-20-40 thresholdcriterion which dictates that an output is only correct if it is greater than 0. ... Back-propagation networks have been used to solvethe problem [12] using a 2-5-5-1 network with shortcut connections between -
A Modular Q-Learning Architecture for Manipulator Task Decomposition…
mi.eng.cam.ac.uk/reports/svr-ftp/auto-pdf/tham_ml94.pdf9 Aug 2005: 010. 2030. 40. 0. 10. 20. 30. 40-0.2. 0. 0.2. 0.4. ... 010. 2030. 40. 0. 10. 20. 30. 40-0.6. -0.4. -0.2. 0. -
THE DEVELOPMENT OF THE CAMBRIDGE UNIVERSITY RT-04 DIARISATION SYSTEM…
mi.eng.cam.ac.uk/reports/full_html/tranter_rt04.html/paper.pdf10 Jan 2005: 19.7spkr-mid 16.7 16.2 23.5 23.5 20.0. ... 40 45 50 55 60 6517.5. 18. 18.5. 19. 19.5. 20. -
Contour-Based Learning for Object Detection Jamie ShottonDepartment…
mi.eng.cam.ac.uk/reports/svr-ftp/shotton_iccv05.pdf8 Aug 2005: 0 10 20 30 40 50. 0.5. 0.6. 0.7. 0.8. 0.9. ... Cipolla. Filteringusing a tree-based estimator. In ICCV03, pages 1063–1070, 2003. [20] J. -
Face Recognition with Image Sets Using Manifold Density Divergence ...
mi.eng.cam.ac.uk/reports/svr-ftp/oa214_CVPR_2005_paper1.pdf8 Aug 2005: 01. 20. 24. 2.5. 2. 1.5. 1. 0.5. 0. 0.5. 1. ... IEEE Conference on Computer Vision and PatternRecognition, pages 313–320, 2003. [20] B. -
��������� �� � �������������� ����������������…
mi.eng.cam.ac.uk/reports/svr-ftp/auto-pdf/sato_tr167.pdf9 Aug 2005: o o;jWvi dUi! ikd. 0 10 20 30 40 50 60 70 80 90-20. ... º> QZ "º>. , //6 7 12;" 8[x'USz:% 6Q / & x 1435 69 x /Y3 F 20 - /UX. -
����������� �� ��������������� ���� � ��� �"!$# #&% ' ...
mi.eng.cam.ac.uk/reports/svr-ftp/auto-pdf/gee_tr174.pdf9 Aug 2005: 10P,= 6 ,Bx| ,=) &,G9o10PB4;,=)20 & &|0S)? ,=E7 C7? 9-x7B ,=q ,W M,G9,W H| Ux77E7 &,Bw!@71(B7 @,=0 & & B4F &S. ... tÖSÉ t Ñ0ÊEÑ&%;Ñ0Ê'#. ( )( ). -100-50. 050. 100. 0. 20. 40. 60. 80. 1002. 4. 6. -
A Face Recognition System for Access Control using Video ...
mi.eng.cam.ac.uk/reports/svr-ftp/arandjelovic_PR06.pdf21 Dec 2005: pus of video sequences of faces in random motion. To learn the model, we took 20. ... For offline learning of illumination subspaces we used 10s video sequences of 20. -
��������� �� �������������� ����� ���� ����� ����! "#��…
mi.eng.cam.ac.uk/reports/svr-ftp/auto-pdf/waterhouse_nips94.pdf9 Aug 2005: 0.8. 1. x. y(x). 0 20. 0.02. 0.04. 0.06. 0.08. x. ... y(x. )f(x. ))2. 0 20. 0.2. 0.4. 0.6. 0.8. 1. x. -
Sparse Finite Element Level-Sets for Anisotropic Boundary Detection…
mi.eng.cam.ac.uk/reports/svr-ftp/weber_scaleSpace2005.pdf24 May 2005: However, the methodology of [13] is not confined to standardsimplices. Here, we obtain a convenient partitioning M of space as Delaunaytetrahedrisation [20] of the vertex-set (2 Z3). -
��������� �� ������� ����������� ������� ��� � �� ��! ...
mi.eng.cam.ac.uk/reports/svr-ftp/auto-pdf/niesler_tr265.pdf9 Aug 2005: 4 5. #. 0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16 0.18 0.20. -
TPAMI0110-0304-1 1..13
mi.eng.cam.ac.uk/reports/svr-ftp/williams_pami2005.pdf20 Apr 2005: the same number of data).Fig. 20 shows the mean CPU utilization of each algorithm. ... 20, no. 10, pp. 1025-1039, Oct. 1998. [11] M. Isard and A. -
STRUCTURAL METADATA RESEARCH IN THE EARS PROGRAM Yang Liu1,5 ...
mi.eng.cam.ac.uk/reports/svr-ftp/tomalin_icassp05.pdf12 May 2005: Additionally, textual information is. 1 2 5 10 20 40 60 80 90 1. ... Woodland, “SU detection forRT-03F at Cambridge University,” http://www.nist.gov/speech/tests/rt/rt2003/fall/presentations/, 2003. [20] Y. -
Learning New Articulator Trajectories for a Speech Production…
mi.eng.cam.ac.uk/reports/svr-ftp/auto-pdf/blackburn_icnn95.pdf9 Aug 2005: 0 10 20 30 40 50 60 70 80 90 100. -
Edge tracking for motion segmentation anddepth ordering P. Smith, ...
mi.eng.cam.ac.uk/reports/svr-ftp/auto-pdf/smith_bmvc1999.pdf9 Aug 2005: The EM iteration process is repeated until convergence (i.e. until there is no significantchange in probabilities), or for 20 iterations. -
��������� �� ���������������������! �"$#&%('�)+*!%�…
mi.eng.cam.ac.uk/reports/svr-ftp/auto-pdf/wu_tr94.pdf9 Aug 2005: 0. 20. 40. 60. 1000 2000 3000 4000. (Hz). (dB). LP(26) Residual. ... 20. 40. 60. 80. 1000 2000 3000 4000. (Hz). (dB). NLP(10-2) Residual Spectrum. -
Sensor Fusion and Occlusion Refinement for Tablet-based AR Georg ...
mi.eng.cam.ac.uk/reports/svr-ftp/klein04tablet.pdf18 Jan 2005: To aid readabil-ity the notation here is loosely based on Welch and Bishop’sexcellent tutorial [20] on the filter. ... 20] G. Welch and G. Bishop. An introduction to the kalman fil-ter. -
Epipolar Geometry from Profiles under Circular Motion Paulo R. ...
mi.eng.cam.ac.uk/reports/svr-ftp/auto-pdf/mendonca_PAMI.pdf9 Aug 2005: Typical values. 20. of lie between -0.5 rad and 0.5 rad, or and. ... perspective projection. Int. Journal of Computer Vision, 31(1):31–50, February 1999. [20] T. -
Desired Utterance Text Extraction of desiredphoneme string andF0…
mi.eng.cam.ac.uk/reports/svr-ftp/auto-pdf/tuerk_eurospeech93a.pdf9 Aug 2005: 09KP B=>#"E UQ9! #EK6O! 0O;"H! (,."I64cc7[V /! 6 =>##D"ELU'. "EK.]. #20 S }B 0l):! ... Nodes 34-43 Position features. Input Vector. Output vector: Node 0 Probability of VoicingNode 1 RMS energyNodes 2-22 20 Cepstral Coefficients. -
SENSORLESS FREEHAND 3DULTRASOUND IN REAL TISSUE: SPECKLE…
mi.eng.cam.ac.uk/reports/svr-ftp/gee_tr510.pdf10 Jan 2005: Even if a practical speckle detector were able to singleout these two patches, we would still have residual distance errors of around 20–30%, and, withjust two offsets, we would be ... Though much improved, the results stillindicate errors of around -
��������� �� �� ������ ��������� ������� �� �!� "$# ...
mi.eng.cam.ac.uk/reports/svr-ftp/auto-pdf/sato_tr177.pdf9 Aug 2005: Linear moments method. Buffon transform method. Proposed method. 0 5 10 15 20 25 30 35 4020. ... Buffon transform method. Proposed method. 0 10 20 30 40 50 60 70 80 9020. -
University of Cambridge, 3D Ultrasound Research Sequential 3D…
mi.eng.cam.ac.uk/reports/svr-ftp/auto-pdf/3dus_tutorial.pdf9 Aug 2005: Sequential 3D Diagnostic Ultrasound using the Stradx System 20. Using voxel arrays. ... In Proceedings of Medical Image Understanding andAnalysis, pages 17–20, Oxford, UK, 1999. -
��������� �� ��������������������������� ���…
mi.eng.cam.ac.uk/reports/svr-ftp/auto-pdf/reinhard_ioa98.pdf9 Aug 2005: $#%#&"'$#%"()!#%" #% ,. 15 20 25 30 35 40 45 50 55 6025. ... 20. 15. 10. 5. 0. 5. 10Initial projection step. first eigenvector depending on τ. -
��������� ��� ��� �������� ��������� � ��…
mi.eng.cam.ac.uk/reports/svr-ftp/auto-pdf/wu_tr124.pdf9 Aug 2005: 600. 700. 800. 900. 1000. 0 10 20 30 40 50 60 70 80 90 100. ... 40. -20. 0. 20. 40. 60. 80. 100. 500 1000 1500 2000 2500 3000 3500 4000. -
3D ELASTOGRAPHYUSING FREEHAND ULTRASOUND J. E. Lindop, G. M. ...
mi.eng.cam.ac.uk/reports/svr-ftp/lindop_tr531.pdf28 Jul 2005: 20]. Models developed by Varghese et al. incorporatedthe effects of electronic noise and signal decorrelation due to axial strain, and they predictedpassbands at 1–10% strain. ... 1G 0.8370 -0.29 1B 0.6250 -0.12. 2G 0.8581 -0.12 2B 0.3641 0.20. -
��������� �� ������� ������������������������ �� ��!� �…
mi.eng.cam.ac.uk/reports/svr-ftp/auto-pdf/burrows_icassp95.pdf9 Aug 2005: Ï. 0 1000 2000 3000 4000 5000 6000 7000 800040. 30. 20. -
��������� �� ��� ��������� ��������� ��� ��� �����! " ...
mi.eng.cam.ac.uk/reports/svr-ftp/auto-pdf/yow_bmvc95.pdf9 Aug 2005: 24350. 510. 1520. 2530. 0. 10. 20. 300.8. 0.6. 0.4. 0.2. ... TF. L. P( R)P(R). 0.85 0.150.20 0.80. TF. R. virtualnode. A. -
Estimating Disparity and Occlusions in Stereo Video Sequences Oliver…
mi.eng.cam.ac.uk/reports/svr-ftp/williams_cvpr2005.pdf9 Aug 2005: 20] J. Sun, H.-Y. Shum, and N.-N. Zheng. Stereo matching usingbelief propagation. -
��������� ��� ������� ��������� ��������� !�"�#��� $�%…
mi.eng.cam.ac.uk/reports/svr-ftp/auto-pdf/tuerk_eurospeech93b.pdf9 Aug 2005: 0 ". ||||zxkxB3x! L x z ||||Zy{ kxK|z! L x z {. RE OEi G0)# Ai'# 20 LE( AFE2 OEiE! L[L[A# 0:O)0:.#7. e-. ... B#A#0 W 0:57'?20:;r 0'! 6!6! /E # /(6'O. 2A;6; D@"Ri! 6p g. -
� ������� �� ��� ���� ����� ��� ����������� ����� ...
mi.eng.cam.ac.uk/reports/svr-ftp/auto-pdf/kim_icslp2000.pdf9 Aug 2005: 0 5 10 15 20 2572. 74. 76. 78. 80. 82. -
����������� ����� ������������� ���������� �����…
mi.eng.cam.ac.uk/reports/svr-ftp/auto-pdf/reinhard_icassp98.pdf9 Aug 2005: 0. 20. 40. 60. Distance monitor along sentence for male and female speaker. -
� ������� �� ���������� ���������������� � �� � ������!"� ...
mi.eng.cam.ac.uk/reports/svr-ftp/auto-pdf/niesler_tr281.pdf9 Aug 2005: 3. 0 10 20 30 40 50 60 70 80 90 1000. ... Consider the first term on the right-hand side of (19) :. X KI I 0 I j X 4 KI I 0 6 j X K I j X (20).
Search history
Recently clicked results
Recently clicked results
Your click history is empty.
Recent searches
Recent searches
Your search history is empty.