Search
Search Funnelback University
1 -
10 of
10
search results for `Download the data` |u:mi.eng.cam.ac.uk
Fully-matching results
-
UNSUPERVISED TRAINING FOR MANDARIN BROADCAST NEWS AND…
mi.eng.cam.ac.uk/research/projects/AGILE/publications/wang_ICASSP07.pdf10 Oct 2007: Acousticsegmentation is performed. The data is then split into wide-bandand narrow-band speech. ... The language models used in these experiments were trainedusing various sources including the LDC giga-word released and theweb download data. -
Practicable Assessment of Cochlear Sizeand Shape from Clinical CT ...
mi.eng.cam.ac.uk/reports/svr-ftp/gee_tr004.pdf27 Jul 2020: The data in Figure 4 is pseudo-clinical CT data, though the method is equally applicable to micro-CT data. ... The first step is to position the template surface at approximately the correct location, by manually iden-tifying three point landmarks in the -
K. Yu, M.J.F. Gales and P.C. Woodland Cambridge University ...
mi.eng.cam.ac.uk/research/projects/AGILE/publications/yu-interspeech07.pdf10 Oct 2007: The procedure used is the same as described in [10]. First,commercials are removed by detecting repeated blocks of audio data.The data is segmented into homogeneous speaker/acoustic conditionblocks and ... The language models used in these experiments -
Predicting hip fracture type with cortical bone mapping (CBM)in ...
mi.eng.cam.ac.uk/reports/svr-ftp/treece_tr695.pdf26 Jan 2015: Receiver operating characteristic (ROC) curves were calculated with leave-one-out cross-validation. Since the extent of the CBM patches were also dependent on the data, SPM for calculating the. ... QCT values were given for a subset of the data, with n = -
Ultrasound attenuation measurementin the presence of scatterer…
mi.eng.cam.ac.uk/reports/svr-ftp/treece_tr502.pdf26 Nov 2004: Such approaches are clearly dependenton a certain level of natural homogeneity in the data, and have the potential to falsely ‘correct’shadows or enhancements where this assumption does not hold. ... In effect, the size of each step gives an -
Imaging the femoral cortex: thickness,density and mass from clinical…
mi.eng.cam.ac.uk/reports/svr-ftp/treece_tr665.pdf29 Apr 2011: cd that best explain the data. The solution, and the idealised data yblur that it implies, areshown in red in the lower panel. ... The former requires just one pass through the data to measure thm, yhm1 and y. -
UNSUPERVISED TRAINING FOR MANDARIN BROADCAST NEWS AND…
mi.eng.cam.ac.uk/~mjfg/wang_ICASSP07.pdf22 Jun 2007: Acousticsegmentation is performed. The data is then split into wide-bandand narrow-band speech. ... The language models used in these experiments were trainedusing various sources including the LDC giga-word released and theweb download data. -
DISCRIMINATIVE SPOKEN LANGUAGE UNDERSTANDINGUSING WORD CONFUSION…
mi.eng.cam.ac.uk/~sjy/papers/hgtt12.pdf20 Feb 2018: Some basic statistics of the corpora are givenin Table 1, and the data is available for download online4. ... For the F-score, a de-gree 2 polynomial was found to model the data best, and for ICE alinear regression was found to be best using F-tests. -
K. Yu, M.J.F. Gales and P.C. Woodland Cambridge University ...
mi.eng.cam.ac.uk/~mjfg/yu-interspeech07.pdf11 Jan 2008: The procedure used is the same as described in [10]. First,commercials are removed by detecting repeated blocks of audio data.The data is segmented into homogeneous speaker/acoustic conditionblocks and ... The language models used in these experiments -
doi:10.1016/j.csl.2004.03.001
mi.eng.cam.ac.uk/~sjy/papers/heyo05.pdf20 Feb 2018: The DARPA Com-municator Travel Data are available to the public as open source download. ... After cleaning up the data, the training set consists of 12,702 utteranceswhile the test set contains 1178 utterances.
Search history
Recently clicked results
Recently clicked results
Your click history is empty.
Recent searches
Recent searches
Your search history is empty.