Search
Search Funnelback University
Did you mean economiespast |u:mi.eng.cam.ac.uk?
101 -
150 of
1,013
search results for Economics test |u:mi.eng.cam.ac.uk
where 12
match all words and 1,001
match some words.
Results that match 1 of 2 words
-
THE CU-HTK MANDARIN BROADCAST NEWS TRANSCRIPTION SYSTEM R. Sinha, ...
mi.eng.cam.ac.uk/~mjfg/sinha_ICASSP06.pdf22 Nov 2006: The finalsystem shows state-of-the-art performance over a range of test sets. ... This approach was not found to perform reliably across differ-ent types of test data. -
Spoken Document Retrieval for TREC-9 at Cambridge University
mi.eng.cam.ac.uk/reports/full_html/johnson_trec9.html/23 Feb 2002: to the query before running the standard BRF technique on the test collection. ... The query expansion was performed on the union of the test and the parallel text collections. -
THE 1998 HTK SYSTEM FOR TRANSCRIPTION OFCONVERSATIONAL TELEPHONE…
mi.eng.cam.ac.uk/reports/svr-ftp/hain_icassp99.pdf27 Sep 2000: VTLN warping in training and test both for GI and GDmodels for full and reduced bandwidth coding. ... 8. ACKNOWLEDGEMENTS. We would like to thank BBN for providing the MiniTrain trainingand MTtest test set definitions. -
C:/SFWDoc/Academic/Publications/2006/ICPR_2006/Final_ContGest/icpr_200…
mi.eng.cam.ac.uk/reports/svr-ftp/sfwong_icpr06a.pdf21 Sep 2006: A detailed report of this test can be found in ourprevious work [10]. ... 20.2 fps).Figure 3 illustrates the recognition process on a typical test-ing clip. -
General Query Expansion Techniques for Spoken Document Retrieval
mi.eng.cam.ac.uk/reports/full_html/jourlin_esca99.html/2 Mar 2000: The experiments reported here use the TREC-7 SDR test data. For this evaluation, the audio documents came from American broadcast radio and TV news programs which had been manually divided ... The five systems described in section 4.3 were evaluated on -
References [1] An inner table retriever for robust table ...
mi.eng.cam.ac.uk/~wjb31/bak.PUBS/3 Nov 2023: Still, there remains a significant performance gap between the English and the cross-lingual test sets. ... One possible approach to this problem uses sub-character decomposition for training and test sentences. -
CAMBRIDGE UNIVERSITY ENGINEERING DEPARTMENT DISCRIMINATIVE…
mi.eng.cam.ac.uk/~mjfg/gales_tr605.pdf13 Aug 2008: In addition tobackground additive noise convolutional distortion was added to test set C. ... For all SNRsand test sets gains over the optimised VTS scheme were obtained. -
Towards Using Conversations with Spoken Dialogue Systems in…
mi.eng.cam.ac.uk/~sjy/papers/lygk16.pdf20 Feb 2018: 1 Introduction. Speaking tests for assessing non-native speakersof English (NNSE) often include tasks involvinginteractive dialogue between a human examinerand a candidate. ... Williams, Kai Yu, Steve Young,and Maxine Eskenazi. 2011. Spoken dialog -
This article appeared in a journal published by Elsevier. ...
mi.eng.cam.ac.uk/~sjy/papers/inyo0920 Feb 2018: Table 3Percent of unseen contexts in the test data. Number of matching features. ... Noduration modification was applied for this test. Twenty subjects participated in the evaluation. -
Spoken Document Retrieval for TREC-8 at Cambridge University
mi.eng.cam.ac.uk/reports/full_html/johnson_trec8.html/30 Mar 2000: This contains the TREC-8 SDR test collection period (Feb 1st to June 30th 1998). ... For development, the 100 hours of TREC-7 SDR test data was used. -
POSTERIOR PROBABILITY DECODING, CONFIDENCE ESTIMATION AND SYSTEM…
mi.eng.cam.ac.uk/reports/full_html/evermann_stw00.html/5 Oct 2000: All experiments were performed on the test sets used in the September 1998 and March 2000 Hub5 evaluations (eval98 and eval00, respectively).. ... It can be seen that for the eval00 test set the unmapped posteriors perform much better than for the eval98 -
tech.dvi
mi.eng.cam.ac.uk/~sjy/papers/bghk13.pdf20 Feb 2018: Reducingthenumber of system components leads to a simpler architecturethat iseasier to implement and test. ... Table 4. Analysis of detected speech segments using V2 and A2models on noisy test setGM2b. -
From Discontinuous To Continuous F0 Modelling In HMM-based…
mi.eng.cam.ac.uk/~sjy/papers/yuty10.pdf20 Feb 2018: Section 4 presents the results of bothobjective and subjective tests. Conclusions then follow. ... 4.2. Subjective listening tests. The previous objective measures show useful comparison infor-mation. -
Speaker Diarisation for Broadcast News
mi.eng.cam.ac.uk/reports/full_html/tranter_odyssey04.html/14 Jun 2004: This was to simulate conditions in the RT-03s evaluation, where there was a temporal gap between the test-audio and the training shows. ... To help minimise the cost of computing the BIC statistics at every point, a faster Hotelling's T test is first -
eps.dis.dur.testa.eps
mi.eng.cam.ac.uk/~mjfg/gales_ASRU09.pdf14 Sep 2010: All three test sets, A, B and C,were used for evaluating the schemes. ... This work used three of the test sets containing digit. sequences (phone numbers). -
Experiments with Fisher Data Gunnar Evermann, Bin Jia, Kai ...
mi.eng.cam.ac.uk/research/projects/EARS/pubs/evermann_sttmay04.pdf25 May 2004: Test setseval03 6h set from Fisher and Swb2-5 data, 72 conversationsdev04 3h set from Fisher, 36 conversations. ... WER on eval03, fsh2004sub models, unadapted, 2003 trigram. Test with adaptation in P1-P2 system:. -
Applying Deep Learning in Non-native Spoken English Assessment
mi.eng.cam.ac.uk/~mjfg/ALTA/presentations/APSIPA2019_Knill.pdf21 Feb 2022: Automate (English) spoken language assessment & learning• without simplifying/limiting form of test: “free speaking”• possibility for richer, interactive, tests• desire to assess communication skills. ... 4/45. Spoken Language Assessment & -
BAYESIAN DIALOGUE SYSTEM FOR THE LET’S GO SPOKEN DIALOGUE ...
mi.eng.cam.ac.uk/~sjy/papers/tykg10.pdf20 Feb 2018: During the control tests, each site taking partin the challenge was asked to find test volunteers. ... Table 2. Overall performance for the Lets Go! spoken dia-logue challenge control tests. -
Speaker Clustering Using Direct Maximisation of the MLLR-Adapted…
mi.eng.cam.ac.uk/reports/full_html/johnson_icslp98.html/8 Mar 2000: For speech transcription problems with widely varying speaker, channel and background conditions, such as broadcast news transcription, it is beneficial to include unsupervised test-data adaptation to individual speakers and data ... The inter-node costs -
UNSUPERVISED TRAINING FOR MANDARIN BROADCAST NEWS AND…
mi.eng.cam.ac.uk/~mjfg/wang_ICASSP07.pdf22 Jun 2007: Table 3 shows the P1-P2 perfor-mance3 of these baseline models on the BC and BN tests sets. ... Table 6. Unadapted %CER onbcmdev05. The models were then evaluated on the BC test set,bcmdev05. -
ADAPTATION OF PRECISION MATRIX MODELS ON LARGE VOCABULARYCONTINUOUS…
mi.eng.cam.ac.uk/reports/svr-ftp/sim_icassp2005.pdf12 Apr 2005: For MLLR mean adaptation, a. SystemAdapt Test Set WER (%)Config eval03 dev04 dev04f. ... This system wasevaluated on botheval03 and dev04 test sets and comparedwith the SATDIAGC system. -
ROBUST DIALOG STATE TRACKING USING DELEXICALISEDRECURRENT NEURAL…
mi.eng.cam.ac.uk/~sjy/papers/htyo14.pdf20 Feb 2018: The set of possi-ble values for each slot was different in the test set, which alsoincluded extra slots. ... A summary of the difference between thesmaller training domain and extended test domain of DSTC3is given in table 1. -
Paper Preparation and Submission Instructions for EUSIPCO 2004
mi.eng.cam.ac.uk/~sjy/papers/poyo06.pdf20 Feb 2018: The aim of this test was to formally evaluate the reduction of breathiness observed in TE speech following glottal source replacement. ... All subjects were naïve raters, unfamiliar with TE speech. Results of the pair-wise perceptual tests are -
Discriminative Classifiers with Generative Kernels for Noise Robust…
mi.eng.cam.ac.uk/~mjfg/gales_INTER08.pdf16 Jun 2008: During recognition the following procedure is used:1. Compensate the acoustic models for the test noise con-. ... dition2. Recognise the test utterance Y to obtain 1-best hypoth-. esis, h = h1,. , -
PROBABLISTIC MODELLING OF F0 IN UNVOICED REGIONS IN HMM ...
mi.eng.cam.ac.uk/~sjy/papers/ytgk09.pdf20 Feb 2018: Altogether 21 listeners, 11 native and 10 non-native,participated in the listening test. ... Statistical significance tests were performed assuming a bino-mial distribution of each choice. -
Two-way Cluster Voting to Improve Speaker Diarisation Performance
mi.eng.cam.ac.uk/reports/full_html/tranter_icassp05.html/31 Mar 2005: NIST,. ''Benchmark Tests : Rich Transcription (RT),''. -
IEEE TRANS. ON ASLP, TO APPEAR, 2011 1 Continuous ...
mi.eng.cam.ac.uk/~sjy/papers/yuyo11.pdf20 Feb 2018: For the test material 30 sentences froma tourist information enquiry application were used. ... This same text materialsas the above preference test were used for all the below. -
paper.dvi
mi.eng.cam.ac.uk/~mjfg/liao_ICASSP07.pdf15 Aug 2007: Thispaper shows that multistyle models benefit from VTS compensationor Joint uncertainty decoding by reducing the mismatch betweentraining and test. ... Figures are reported against thedev03 test set encompassing3 hours of shows from six different news -
Generating and Evaluating Segmentations for Automatic Speech…
mi.eng.cam.ac.uk/reports/full_html/tranter_icassp04.html/13 Jul 2004: NIST,. Benchmark Tests : Rich Transcription (RT). -
Class-based language model adaptation using mixtures of word-class…
mi.eng.cam.ac.uk/reports/full_html/moore_icslp00.html/2 Nov 2000: The full HUB-4 1997 evaluation test set was used to assess language model performance. ... Treating this as the baseline result to improve upon, statistical significance was tested by using the Wilcoxon signed-rank test at a 5% significance level.. -
Tue.P5b.03 Model-based Approaches for Degraded Channel Modelling in…
mi.eng.cam.ac.uk/~mjfg/gales_is2012.pdf13 Jun 2013: As the test data was Levantine Arabic,there is little appropriate data for training language models. ... Forthis work all the clean Levantine Arabic transcriptions, exclud-ing the dev1 test data, was used. -
MultiMedia Document Retrieval (1997-2000) - Progress
mi.eng.cam.ac.uk/research/projects/Multimedia_Document_Retrieval/progress.html7 Oct 2001: The final HTK system designed from the results of these experiments yielded an overall word error rate of 22.0% on the 1996 unpartitioned broadcast news development test data and just ... but if the (small) test collection is used, then average precision -
Cross-Lingual Spoken Language Understanding from Unaligned Data…
mi.eng.cam.ac.uk/~sjy/papers/lemy10.pdf20 Feb 2018: 1. English STC baseline tested on the English test set asreported in [6]. ... Table 2: BLEU scores [?] over the test set translated usingGoogle Translate and Moses. -
Machine Learning of Level and Progression in Spoken EAL ...
mi.eng.cam.ac.uk/~mjfg/ALTA/presentations/CEP_Feb2016_Knill.pdf21 Feb 2022: Business Language Testing Service (BULATS) Spoken Tests. • Example of a test of communication skills A. ... Use automatic grader • for grading practice tests/learning process • in combination with human graders. • -
The 1998 HTK Broadcast News Transcription System: Development and…
mi.eng.cam.ac.uk/reports/full_html/woodland_darpa99.html/2 Mar 2000: During system development the BNeval97 test set was used. 4.1 Increased Acoustic Training Data. ... All these sources excluded data from the designated test epochs. This corpus was denoted LMtrain98. -
Kim et al.: 2003 CU-HTK BN-E Systems 2003 CU-HTK ...
mi.eng.cam.ac.uk/research/projects/EARS/pubs/kim_rt03s.pdf24 Jul 2003: Baseline Acoustic Models: Results (I). • Development test setsBNeval98 two 1.5-hour data setsBNeval02 1-hour data setBNdev03 three hours of TDT-4 data from Jan ’01 transcribed by STT sites. • ... MLE baseline system– Add 3rd derivatives HLDA, -
nips7.dvi
mi.eng.cam.ac.uk/reports/svr-ftp/williams_nips2004.pdf9 Aug 2005: This is restrictive in practicein that test data may contain distortions that take it outside the strict ambit of the trainingpositives. ... This can be computed in terms of likelihoods. (1)so then the test becomes (2)where. -
A Clustering Approach to Semantic Decoding Hui Ye and ...
mi.eng.cam.ac.uk/~sjy/papers/yeyo06b20 Feb 2018: There were in total47 distinct semantic slots. To parse a test sentence, it was first preprocessed to apply thesame lexical class substitution as used in training, then aligned toeach of the ... 4.2. Results. The Y-clustering approach was tested using -
The Cambridge University March 2005 Speaker Diarisation System R. ...
mi.eng.cam.ac.uk/reports/svr-ftp/sinha_eurospeech05.pdf23 Jun 2005: data when processing all the test data togetheror 8.6% when processing the test data shows independently. ... The previous experiment was repeated adding the test dataitself, E, instead of the dev data. -
Unsupervised Language Model Adaptation for Mandarin…
mi.eng.cam.ac.uk/reports/svr-ftp/mrva_icslp06.pdf20 Jan 2007: All the test sets comprise ofshows where a show is a broadcast of a TV or radio program. ... 3.4. Results. The tests were carried out on a BC test set (dev05bcm: 3 hours,30k words, 5 shows) and two BN test sets (eval04: 1.3 hours, 11kwords, 3 -
Recent Improvementsin the CUED CTS SU System M. Tomalin ...
mi.eng.cam.ac.uk/research/projects/EARS/pubs/tomalin_mar05mde.pdf28 Apr 2005: increasing mixtures.• Obtain prob from each SU subtype GMM for each feature vector in test data.• Place GMM probs on arcs of lattice and decode as usual. ... Initial GMM results for dev03f test set:. SYSTEM DEL INS SUBS %Err. -
Version IB/2 3 (a) This network follows a typical ...
mi.eng.cam.ac.uk/~cipolla/lectures/4F12/Examples/exam/4F12_2023_CRIB.pdf23 Oct 2023: B = 1, otherwise. is a margin parameter. At test time the a single image is used to obtain its embedding vector and find thenearest neighbour in the database of the embedding ... Only the the deconvolution layer parameters are trained. Test use. At test -
ITERATIVE UNSUPERVISED ADAPTATION USING MAXIMUMLIKELIHOOD LINEAR…
mi.eng.cam.ac.uk/reports/svr-ftp/woodland_icslp96.pdf17 Oct 2000: standard clean modelset and then to perform further adaptation for the actual test data [7]. ... For the evaluation test H3-P0 data one microphone was selectedby NIST for each speaker. -
��������� �� ������� ����������� ������� ��� � �� ��! ...
mi.eng.cam.ac.uk/reports/svr-ftp/auto-pdf/niesler_tr265.pdf9 Aug 2005: The test-set wassubjected to the same preprocessing steps as the training corpus. ... The OOV rate for the describedvocabulary is 0.62% on the test-set. -
Some Results on CTS Quick Transcription andFisher Data Phil ...
mi.eng.cam.ac.uk/research/projects/EARS/pubs/woodland_martigny03.pdf16 Sep 2003: 357 female sides (23.75hrs), only 142 male sides (9.43hrs). • Same test setup as for 20hr QT expts. ... MPE gain is larger than MLE gain. • Gains larger on female test data (biased training). -
ADAPTIVE TRAINING USING STRUCTURED TRANSFORMS
mi.eng.cam.ac.uk/reports/svr-ftp/kai_icassp04st.pdf27 May 2004: During training and test adaptation, a global in-terpolation weight was estimated and separate speech and silencetransforms were used for CMLLR. ... Using ST for test adaptation again yieldedabout the same as a GD system with CMLLR. -
Real user evaluation of spoken dialogue systems using Amazon ...
mi.eng.cam.ac.uk/~sjy/papers/jkgm11.pdf20 Feb 2018: Two tailed z-tests at the 95% confidence levelwere used to test the statistical significance of the differencebetween the success rates. ... The use of the AMT workers is substantially more efficientwhen compared to a controlled test. -
paper.dvi
mi.eng.cam.ac.uk/~mjfg/richter_EURO99.pdf19 Nov 2010: The two forms of modified tail distribution modelling were inves-tigated on the 1998 Hub4 partitioned evaluation test set. ... The test was performed on a subset of the 1997 par-titioned evaluation that was used for development [4]. -
Applying Deep Learning in Non-native Spoken English Assessment
mi.eng.cam.ac.uk/~kmk/presentations/APSIPA2019_Knill_Keynote.pdf21 Nov 2019: Automate (English) spoken language assessment & learning• without simplifying/limiting form of test: “free speaking”• possibility for richer, interactive, tests• desire to assess communication skills. ... 4/45. Spoken Language Assessment & -
Professor Steve Young, Publications
mi.eng.cam.ac.uk/~sjy/publications.htmlYu, S. Young, M. Eskenazi (2011). "Spoken Dialog Challenge 2010: Comparison of Live and Control Test Results. "
Refine your results
Date
- 422 2018
- 87 2005
- 51 2006
- 48 2000
- 45 2020
- 30 2008
- 30 2004
- 30 2022
- 26 2003
- 25 2007
- 25 2023
- 23 2019
- 20 2014
- 19 Past year
- 18 2015
- 18 2021
- 16 2013
- 15 2009
- 14 2016
- 14 2010
- 14 2024
- 9 2012
- 8 2017
- 7 2011
- 7 2001
- 6 Past 3 months
- 6 Past 6 months
- 5 2002
- 4 Uncertain
- 2 1999
- 2 Past fortnight
- 2 Past month
- 1 Past week
- 1 Yesterday
Search history
Recently clicked results
Recently clicked results
Your click history is empty.
Recent searches
- `download and complete our` (1) · moments ago
Recent searches
Your search history is empty.