Search
Search Funnelback University
- Refined by:
- Date: 2009
1 -
10 of
20
search results for TALK:PC53 20 |u:mi.eng.cam.ac.uk
where 0
match all words and 20
match some words.
Results that match 1 of 2 words
-
Sequence Kernels for Speaker and SpeechRecognition Mark Gales - ...
mi.eng.cam.ac.uk/~mjfg/jhu09.pdf16 Jul 2009: Cambridge UniversityEngineering Department. JHU Workshop 2009 20. Sequence Kernels for Speaker and Speech Recognition. ... 20, pp. 210–229, 2005. [3] H. Lodhi, C. Saunders, J. Shawe-Taylor, N. -
paper_pcw_v1.dvi
mi.eng.cam.ac.uk/research/projects/AGILE/publications/park_interspeech09.pdf7 Oct 2009: MPE PLP Original 20.5 22.8PLPMLP1350hr Fast 17.7 20.0. Semi-fast 17.7 19.9Rebuild 17.7 19.9. ... SystemWER. dev07 eval07 dev08. ML. PLP 21.1 22.9 25.1MLP1350hr 20.2 21.2 22.6PLPMLP200hr 18.7 20.1 21.6PLPMLP1350hr 18.6 19.8 21.2PLPMLP1350hr big -
paramStudy_V_(TechR).dvi
mi.eng.cam.ac.uk/reports/svr-ftp/Shin_TR600.pdf4 Aug 2009: 15. 20. 25. Lateral, mm. ISNR = 2.04 dB (log: 1.32 dB). ... 15. 20. 25. Lateral, mm. ISNR = 2.40 dB (log: 1.38 dB). -
Learning to Track with Multiple Observers Björn StengerComputer…
mi.eng.cam.ac.uk/~cipolla/publications/inproceedings/2009-CVPR-hand-trackingpdf28 Jul 2009: and subsequently combinetheir output, by either switching between them [2] or byprobabilistically merging them [8, 15, 18, 20]. ... 20] P. Pérez, J. Vermaak, and A. Blake. Data fusion for visual track-ing with particles. -
MULTIPLE KERNEL LEARNING FOR SPEAKER VERIFICATION C. Longworth and ...
mi.eng.cam.ac.uk/~mjfg/cl336_ICASSP08.pdf2 Mar 2009: 0.1 0.2 0.5 1 2 5 10 20 40. 0.1. 0.2. ... 0.5. 1. 2. 5. 10. 20. 40. False Alarm probability (in %). Mis. s pr. obab. ility. (in. %). GMMLLR. 128. λ128. λ. -
3D ULTRASONIC STRAIN IMAGINGUSING FREEHAND SCANNING ANDA…
mi.eng.cam.ac.uk/reports/svr-ftp/housden_tr623.pdf28 Jan 2009: 1. depends partly on the size and shape of the compressor [20, 23]. ... The lateral and elevationalfocal depths are both 20 mm. 3.1 Mechanical stress artefacts. -
A Generalised Derivative Kernel for Speaker Verification C. Longworth …
mi.eng.cam.ac.uk/~mjfg/cl336_INTER08.pdf2 Mar 2009: 9.15 10.38 11.11Approx 12.21 10.20 9.25 10.29. -
report.dvi
mi.eng.cam.ac.uk/reports/svr-ftp/ijaz_tr635.pdf20 Aug 2009: 4. Using normalised mutualinformation, the Nelder-Mead simplex algorithm converged in all cases out of which 80% requiredno restarts and 20% required one restart. ... Ultrasound in Medicine and Biology, 20(4):923–936, 1994. [7] D. F. Leotta, P. -
Model-Based Approaches to Speaker andEnvironment Adaptation Mark…
mi.eng.cam.ac.uk/~mjfg/mjfg_china09.pdf28 Apr 2009: s)d = Adb. (s)ml bd. Cambridge UniversityEngineering Department. Tsinghua University April 2009 20. ... Rep. TR611, February2009, available from http://mi.eng.cam.ac.uk/mjfg/. [20] R. Gopinath, B. Ramabhadran, and S. -
EM_spd_D4Tech.dvi
mi.eng.cam.ac.uk/reports/svr-ftp/Shin_TR626.pdf4 Aug 2009: Numerous. methods have been proposed [19, 20, 21, 22]. They differ from one another in how they estimate. ... 0.5. 1. 1.5. Err. or,. %. Lateral focus, mm. 9.20. mm.
Search history
Recently clicked results
Recently clicked results
Your click history is empty.
Recent searches
Recent searches
Your search history is empty.