Search
Search Funnelback University
- Refined by:
- Date: 2018
1 -
50 of
600
search results for TALK:PC53 20 |u:mi.eng.cam.ac.uk
where 0
match all words and 600
match some words.
Results that match 1 of 2 words
-
JournalPaperDRAFTV0.20
mi.eng.cam.ac.uk/~sjy/papers/wipy05a.pdf20 Feb 2018: 18 4.1 MDP BASELINE. 20 4.2 RESULTS & DISCUSSION. 22. 5 EVALUATING & IMPROVING A HANDCRAFTED POLICY. ... The MDP State Estimator is updated using a deterministic function:. ),,( MDPSESSE aosfs SE ′=′ (20). -
34 1 2 Localaffine deformation Total affinedeformation Robot…
mi.eng.cam.ac.uk/~cipolla/publications/article/2000-IJCV-visual-servo.pdf13 Mar 2018: 000000. 0"/20. 0. 00#$0. =. 0. $#$0000. 00. ( not! )0#"/200. ... 0 5 10 15 20 25 30 35. "fourier.txt". 0. 0.02. -
Simplifying very deep convolutional neural network architectures for…
mi.eng.cam.ac.uk/UKSpeech2017/posters/j_rownicka.pdf3 Jul 2018: training set of Aurora4. Model A B C D AVGDNN/clntr 2.71 43.00 24.06 58.66 45.48VDCNN-max-4FC/clntr 2.32 35.99 21.20 ... 53 39.62VDCNN-allconv/clntr 1.98 40.62 20.08 55.57 42.80. -
a) Rotation b) Isotropic Expansion c) Pure Shear 0 ...
mi.eng.cam.ac.uk/~cipolla/publications/inproceedings/1996-BMVC-Cross.pdf13 Mar 2018: 17. 17.5. 18. 18.5. 19. 19.5. 20. 20.5. Sample Number. Tim. -
34 1 2 Localaffine deformation Total affinedeformation Robot…
mi.eng.cam.ac.uk/~cipolla/archive/Publications/article/2000-IJCV-visual-servo.pdf13 Mar 2018: 000000. 0"/20. 0. 00#$0. =. 0. $#$0000. 00. ( not! )0#"/200. ... 0 5 10 15 20 25 30 35. "fourier.txt". 0. 0.02. -
The Linear Transformation of LF Glottal Waveforms for Voice ...
mi.eng.cam.ac.uk/~sjy/papers/poyo08b20 Feb 2018: 0.5. 1. 0 5 10 15 20 25 30 35 40 450. ... 0.1. 0.2. 0 5 10 15 20 25 30 35 40 450. -
a) Rotation b) Isotropic Expansion c) Pure Shear 0 ...
mi.eng.cam.ac.uk/~cipolla/archive/Publications/inproceedings/1996-BMVC-Cross.pdf13 Mar 2018: 17. 17.5. 18. 18.5. 19. 19.5. 20. 20.5. Sample Number. Tim. -
Reward-Balancing for Statistical Spoken Dialogue Systems…
mi.eng.cam.ac.uk/~sjy/papers/ubcy17.pdf20 Feb 2018: 0. 5. 10. 15. 20. 0.0. 0.2. 0.4. 0.6. 0.8. 1.0. ... domains comparing the baseline (rws = 20, w = (0.5,0.5)) with the optimised balance. -
inihsampling point Image edge Active Contour Affine Stereo System ...
mi.eng.cam.ac.uk/~cipolla/publications/contributionToEditedBook/1994-CUP-uncalibrated.pdf13 Mar 2018: Controller. Controller. coordscoordscoords. 0. 0.02. 0.04. 0.06. 0.08. 0.1. 0.12. 0 5 10 15 20 25. ... derr.nonoise.h.affine". 0. 0.02. 0.04. 0.06. 0.08. 0.1. 0.12. 0 5 10 15 20 25. -
Deep Neural Network Approach for theDialog State Tracking Challenge…
mi.eng.cam.ac.uk/~sjy/papers/hety13.pdf20 Feb 2018: Structure[] 0.6751 0.8074 0.5658. [50] 0.6679 0.8046 0.5450[20] 0.6656 0.8060 0.5394. ... Training used T =10, the full feature set,a [20, 10, 2] hidden structure and the shared ini-tialisation training method. -
INDICATOR VARIABLE DEPENDENT OUTPUT PROBABILITY MODELLING…
mi.eng.cam.ac.uk/~sjy/papers/tuyo01.pdf20 Feb 2018: avg. F0 F1 F2 F3 F4 F5 FX12mix base 21.8 12.4 20.5 31.5 31.5 24.3 27.5 43.924mix base 21.3 12.2 ... relabelled training datasum 21.4 12.5 20.1 30.2 31.2 24.7 25.1 42.2full 21.4 12.5 20.1 30.6 31.0 24.6 -
Automatic Telephone Voice Analysis and TherapyLadan Baghai-Ravary and …
mi.eng.cam.ac.uk/UKSpeech2017/posters/l_baghai-ravary_poster.pdf24 Jan 2018: The measures include:. CalibrationThe VoiScan parameters for an individual can be compared with population statistics which have been calculated from 20,000 telephone calls in 12 countries. -
User study of the Bayesian Update of Dialogue State ...
mi.eng.cam.ac.uk/~sjy/papers/tgkm0820 Feb 2018: 2) was chosen. 0 20 40 60 80 1000. 0.2. 0.4. ... 0.6. 0.8. 1MDPHDC. 0 20 40 60 80 1000. 0.2. 0.4. -
yeyo05b.dvi
mi.eng.cam.ac.uk/~sjy/papers/yeyo05b.pdf20 Feb 2018: Native: Chinese sentences 11-20 and 31-40 spoken by all 20Chinese speakers, 400 utterances in total. ... NonNativeAB: Chinese sentences 11-20 and 31-40 spoken byall 20 English speakers, 400 utterances in total. -
ICSLPDataCollection-10
mi.eng.cam.ac.uk/~sjy/papers/wiyo04b.pdf20 Feb 2018: 0. 5. 10. 15. 20. 25. 30. 35. Hi Med Low NoneWER Target. ... 0%. 20%. 40%. 60%. 80%. 100%. REPAIR ASKQ REPAIR ASKQ. Hi MedStrategy & WER Target. % -
inihsampling point Image edge Active Contour Affine Stereo System ...
mi.eng.cam.ac.uk/~cipolla/archive/Publications/contributionToEditedBook/1994-CUP-uncalibrated.pdf13 Mar 2018: Controller. Controller. coordscoordscoords. 0. 0.02. 0.04. 0.06. 0.08. 0.1. 0.12. 0 5 10 15 20 25. ... derr.nonoise.h.affine". 0. 0.02. 0.04. 0.06. 0.08. 0.1. 0.12. 0 5 10 15 20 25. -
k-Nearest Neighbor Monte-Carlo Control Algorithmfor POMDP-based…
mi.eng.cam.ac.uk/~sjy/papers/lgjk09.pdf20 Feb 2018: 0 10 20 30 40 50. Succ. essf. ul C. ompl. ... 7. 8. 9. 10. 11. 12. 13. 14. 0 10 20 30 40 50. -
Toward Multi-domain Language Generation usingRecurrent Neural…
mi.eng.cam.ac.uk/~sjy/papers/wgmr15.pdf20 Feb 2018: During decoding, we over-generated 20 utterances and selected the top 5 realisations for each DAaccording to the following reranking criteria,. ... Proceedings of the IEEE,1990. [20] Kishore Papineni, Salim Roukos, Todd Ward, and Wei-Jing Zhu. -
Reward Shaping with Recurrent Neural Networks for Speeding upOn-Line…
mi.eng.cam.ac.uk/~sjy/papers/svgm15.pdf20 Feb 2018: Per turn reward was set to -1and final reward 20 for task success else 0. ... In Proc of the IEEE, vol-ume 99, pages 1–20. A Training Constraint/Loss Function. -
Emotion Conversion using F0 Segment Selection Zeynep Inanoglu, Steve…
mi.eng.cam.ac.uk/~sjy/papers/inyo08.pdf20 Feb 2018: 20% 19.3% 60.7%. 21.3% 47.3% 31.3%. Figure 2: Preference scores for each F0 prediction method andemotion. ... Angry Surprised Sad Can’t decideAngry 99.3% 0.7% 0% 0%. Surprised 20.0% 66.0% 0% 14.0%Sad 0.7% 0% 96.0% 3.3%. -
Using Wizard-of-Oz simulations to bootstrap…
mi.eng.cam.ac.uk/~sjy/papers/wiyo03.pdf20 Feb 2018: 6 Discussion. The majority of the data collected was “usable”: although 26.7% of turns were excluded, 20.5% of these were due to a well-defined task not under study ... 20.0%. 30.0%. 40.0%. 50.0%. 60.0%. 70.0%. 80.0%. 0 500 1000 1500 2000 -
acl2010.dvi
mi.eng.cam.ac.uk/~sjy/papers/gjkm10.pdf20 Feb 2018: 1. 20 60 100 140 180 220 260 300 340 380 420 460 500 540 580 620. ... 50. 45. 40. 35. 30. 25. 20. 15. 10. 5. 0. -
Intonation Modelling and Adaptation for Emotional Prosody Generation…
mi.eng.cam.ac.uk/~sjy/papers/inyo0520 Feb 2018: When analyzed on a per listener basis, the distribution of preferences in partone (happy) suggest strong bimodality: eight of the fourteen listeners had a meanidentification rate of 5 out of 20, ... Proc. of Speech Communication (20 and 03) vol.40 -
ON-LINE POLICY OPTIMISATION OF BAYESIAN SPOKEN DIALOGUE SYSTEMS…
mi.eng.cam.ac.uk/~sjy/papers/gbhk13.pdf20 Feb 2018: Thismethod can be elegantly incorporated into the on-line GP-Sarsa al-gorithm [20]. ... The reward function was set to give a reward of 20 for success-ful dialogues, zero otherwise. -
POLICY COMMITTEE FOR ADAPTATION IN MULTI-DOMAIN SPOKEN…
mi.eng.cam.ac.uk/~sjy/papers/gmsv15.pdf20 Feb 2018: The reward function allocates 1 at each turnto encourage shorter dialogues, plus 20 at the end of each suc-cessful dialogue. ... 20] S Keizer, M Gašić, F Jurčı́ček, F Mairesse, B Thomson,K Yu, and S Young, “Parameter estimation for -
PHONETIC AND GRAPHEMIC SYSTEMS FOR MULTI-GENRE BROADCASTTRANSCRIPTION …
mi.eng.cam.ac.uk/~ar527/wang_icassp2018.pdf4 Nov 2018: This serves as the basis for varioussystem combination approaches [18, 19, 20, 21]. ... the systems being combined [20]:. Ŵ = argminW. {Mm=1. λmW′H. P(W′|O1:T ; Φm. -
Dialogue Context Sensitive Speech Synthesis using Factorized Decision …
mi.eng.cam.ac.uk/~sjy/papers/tgbh14.pdf20 Feb 2018: A set of 20 previously collected dialogues were randomlyselected for the restaurant domain. ... 90, no. 9, pp.1406–1413, 2007. [20] K. Yu, H. Zen, F. -
Evaluating semantic-level confidence scores with multiple hypotheses…
mi.eng.cam.ac.uk/~sjy/papers/tygk08.pdf20 Feb 2018: 5 19.2 12.2 0.212 1.10335.3 InfEv 100 20.6 7.9 — 0.941. ... 0 10 20 30 40 500. 1. 2. 3. 4. Oracle Error Rate. -
Learning from Real Users: Rating Dialogue Success with Neural ...
mi.eng.cam.ac.uk/~sjy/papers/svgk15.pdf20 Feb 2018: 0. This hard label is used to determine whether to give afinal reward of 20 during policy learning, as per Eqn. ... All policies are trained by GP-SARSA [4] andthe summary action space contains 20 actions. -
ERROR SIMULATION FOR TRAINING STATISTICAL DIALOGUE SYSTEMS Jost…
mi.eng.cam.ac.uk/~sjy/papers/scty07b20 Feb 2018: A computationally less expensive word-level error simulationmethod for training statistical dialogue systems has been suggestedby [20]. ... 20] O. Pietquin and T. Dutoit, “A probabilistic framework for dia-log simulation and optimal strategy -
DISTRIBUTED DIALOGUE POLICIES FOR MULTI-DOMAIN STATISTICAL…
mi.eng.cam.ac.uk/~sjy/papers/gkty15.pdf20 Feb 2018: allocates 1 at each turn to encourage shorterdialogues, plus 20 at the end of each successful dialogue. ... Each user was assigned specific tasksin the SFRestaurant subdomain and then asked to call the system ina similar set-up to that described in [20, -
Tsiakoulis_Pirros_1293
mi.eng.cam.ac.uk/~sjy/papers/tghp12.pdf20 Feb 2018: About 25 subjects were asked to perform a set of 20 randomly generated tasks. ... errors. 0 10 20 30 40 50 60. 0.5. 0.6. 0.7. 0.8. -
Challenges for AI in Spoken Communication March 2017 Dr ...
mi.eng.cam.ac.uk/~kmk/presentations/CUED_Knill_20170328_enc.pdf22 Jan 2018: Unless otherwise indicated,the LSTM is unrolled for 20 time steps for training with truncatedbackpropagation through time (BPTT). ... The LSTM is then unrolled for 20timesteps, and thus consumes a larger context of 20 l. -
Active Memory Networks for Language Modeling O. Chen, A. ...
mi.eng.cam.ac.uk/~ar527/chen_is2018.pdf15 Jun 2018: All models were trained with roughly 16–20 million param-eters to perform a fair comparison. ... 6, no. 2, pp. 181–214, 1994. [20] Y. Wang, X. Chen, M. -
is-05-hvs6_final
mi.eng.cam.ac.uk/~sjy/papers/seyo05.pdf20 Feb 2018: 52 class n-gram 26.3 25.0 24.9 HVS_52 21.7 20.4 20.1. Table 1: Perplexity for models of varied stack depths trained for 250 iterations. -
Online_ASRU11.dvi
mi.eng.cam.ac.uk/~sjy/papers/gjty11.pdf20 Feb 2018: TABLE IVCOMPARISON BETWEEN WORD ERROR RATES FOR DIFFERENT CORPORA. Word error rateSimulator trained policy 20.85Online learning 22.93Random policy 25.22. ... 10. 20. 30. 40. 50. 60. 70. 80. 90. 100. Word error rate. -
poyosp08
mi.eng.cam.ac.uk/~sjy/papers/dpyo08.pdf20 Feb 2018: Subjective evaluation. A perceptual test using a panel of 20 listeners was carried out. ... T - O 0.66 0.20 0.14. Table 8: Preference test results. -
A DATA-DRIVEN SPOKEN LANGUAGE UNDERSTANDING SYSTEM Yulan He Steve ...
mi.eng.cam.ac.uk/~sjy/papers/heyo03b.pdf20 Feb 2018: 0 5 10 15 20 252.2. 2.7. 3.2. 3.7. 4.2. Hypothesis recognized list size (N in Nbest) for NOV93. ... 20] K. Hacioglu and W. Ward, “Combining language models:Oracle approach,” in Human Language Technology Confer-ence, San Diego, CA, Mar. -
DISTRIBUTED DIALOGUE POLICIES FOR MULTI-DOMAIN STATISTICAL…
mi.eng.cam.ac.uk/~sjy/papers/gkty15a.pdf20 Feb 2018: allocates 1 at each turn to encourage shorterdialogues, plus 20 at the end of each successful dialogue. ... Each user was assigned specific tasksin the SFRestaurant subdomain and then asked to call the system ina similar set-up to that described in [20, -
Scaling Up POMDPs for Dialog Management: The “Summary POMDP” ...
mi.eng.cam.ac.uk/~sjy/papers/wiyo05.pdf20 Feb 2018: 2 To form a baseline, 20,000 dialog turns were then run with the. ... The summary POMDP method was then applied using 20,000 dialog turns, 50 belief points and 50 iterations. -
14 INGENIA INGENIA ISSUE 54 MARCH 2013 15 Talking ...
mi.eng.cam.ac.uk/~sjy/papers/youn13.pdf20 Feb 2018: Typical word error rates are now between 3 and 8% for clean, carefully spoken speech, rising to 20% or greater for conversational speech in noisy environments. ... XXXXXXTALKING TO MACHINES. INGENIA ISSUE 54 MARCH 2013 21 20 INGENIA. -
Uncertainty management for on-line optimisation of a…
mi.eng.cam.ac.uk/~sjy/papers/dgcg11.pdf20 Feb 2018: A positive reward of 20 isgiven at the end of the dialogue if the DM managed to fulfilthe user request and a penalty of 1 is applied per system turnto encourage ... The new exploration schemes were evaluated in this paperusing GP-SARSA. In the future, -
Real user evaluation of spoken dialogue systems using Amazon ...
mi.eng.cam.ac.uk/~sjy/papers/jkgm11.pdf20 Feb 2018: complete at least 20 dialoguesduring the session, under continuous supervision of a researchteam member. ... Trial # users average # calls median # callsAMT 140 6.5 2Cambridge 17 24.4 20. -
Natural Belief-Critic: a reinforcement algorithm for parameter…
mi.eng.cam.ac.uk/~sjy/papers/jtkm10.pdf20 Feb 2018: The final system was evaluated as in the firsttask; although in this case, the training and evaluation procedurewas executed 20 times. ... by 10%-20% in comparison with the rewards obtained whenusing the handcrafted parameters. -
naacl.dvi
mi.eng.cam.ac.uk/~sjy/papers/stwy07.pdf20 Feb 2018: Usergoals are randomly generated and an (arbitrary) rewardfunction assigning 20 points for successful completionand -1 for every dialogue turn is used. -
Evaluation of Statistical POMDP-basedDialogue Systems in Noisy…
mi.eng.cam.ac.uk/~sjy/papers/ybgh14.pdf20 Feb 2018: This isdone by letting the dialogue system interact with a user simulator[20]. ... vol. 7, no. 3, 2011. 20. J. Schatzmann, B. Thomson, K. Weilhammer, H. -
SPOKEN LANGUAGE UNDERSTANDING FROM UNALIGNED DATA USINGDISCRIMINATIVE …
mi.eng.cam.ac.uk/~sjy/papers/mgjk09.pdf20 Feb 2018: by specifying a VoiceXML grammarthat instantiates slot values. Within the past 20 years, research on spoken language under-standing (SLU) has produced models that alleviate these issuesby learning to derive a ... 94.20 95.26TownInfo dataset with ASR -
DISCRIMINATIVE SPOKEN LANGUAGE UNDERSTANDINGUSING WORD CONFUSION…
mi.eng.cam.ac.uk/~sjy/papers/hgtt12.pdf20 Feb 2018: F-score. C.Net. ContextPhoenix. Predicted F-score. 0 20 40 60 80 100. ... 20,no. 4, pp. 495–514, Oct. 2006. [20] M. Gašić, P. Tsiakoulis, M. -
POLICY OPTIMISATION OF POMDP-BASED DIALOGUE SYSTEMS WITHOUT…
mi.eng.cam.ac.uk/~sjy/papers/ghtt12.pdf20 Feb 2018: It can be shown [20] that the marginal likelihood of the observedrewards is modelled by. ... 20] M Gašić, Statistical Dialogue Modelling, PhD thesis, Univer-sity of Cambridge, 2011. -
sigdial11_sdc10-Feb27-V2
mi.eng.cam.ac.uk/~sjy/papers/bbch11.pdf20 Feb 2018: Table 4: Average dialogue WER. 0 20 40 60 80 100. ... WER for the three fully tested systems. 0 20 40 60 80 100.
Search history
Recently clicked results
Recently clicked results
Your click history is empty.
Recent searches
Recent searches
Your search history is empty.