Search
Search Funnelback University
- Refined by:
- Date: 2018
1 -
10 of
222
search results for KaKaoTalk:vb20 200 |u:mi.eng.cam.ac.uk
where 0
match all words and 222
match some words.
Results that match 1 of 2 words
-
RTSC MAGPHASE VOCODER: MAGNITUDE ANDPHASE ANALYSIS/SYNTHESIS FOR…
mi.eng.cam.ac.uk/UKSpeech2017/posters/f_espic.pdf19 Jan 2018: rate thanthe 200 frames-per-second typical in many sys-tems. -
A HIERARCHICAL ATTENTION BASED MODEL FOR OFF-TOPIC SPONTANEOUSSPOKEN…
mi.eng.cam.ac.uk/~mjfg/ALTA/publications/ASRU2017/HierarchicalAttentionBased/hierarchical-attention-based.pdf24 Jan 2018: Both models were implemented in Tensorflow [18] and contain two400 dimensional BiLSTM encoders with TanH non-linearities, 200for the forward states and 200 for the backward states. ... The binary classifier was a DNN with 2 hidden lay-ers of 200 -
Intonation Modelling and Adaptation for Emotional Prosody Generation…
mi.eng.cam.ac.uk/~sjy/papers/inyo0520 Feb 2018: 0.6 0.8 1 1.2 1.4 1.6 1.8 2150. 200. 250. f0. ... 39 39.5 40 40.5 41 41.5 42140. 160. 180. 200. 220. -
Incremental on-line adaptation of POMDP-based dialogue managers…
mi.eng.cam.ac.uk/~sjy/papers/gktb14.pdf20 Feb 2018: Bearing in mind that the initial parts of. 1The performance deteriorated after 200 dialogues due to an in-crease in speech understanding errors. ... Table 2: Word error rate for different domainsDomain #Adaptation diags #Diags WERSFCore 200 399 15SF1Ext -
Using Wizard-of-Oz simulations to bootstrap…
mi.eng.cam.ac.uk/~sjy/papers/wiyo03.pdf20 Feb 2018: 0. 50. 100. 150. 200. 1 2 3 4 5 6 7 8 9 10 11 12. ... basis. 0. 50. 100. 150. 200. 1 5 9 13 17 21 25 29 33. -
Multi-domain Dialog State Tracking using Recurrent Neural Networks…
mi.eng.cam.ac.uk/~sjy/papers/mstg15.pdf20 Feb 2018: 0 200 400 600 800. 50. 55. 60. 65. In-domain InitialisationOut-of-domain Initialisation. ... 0 200 400 600 800. 40. 60. 80. In-domain InitialisationOut-of-domain Initialisation. -
Uncertainty management for on-line optimisation of a…
mi.eng.cam.ac.uk/~sjy/papers/dgcg11.pdf20 Feb 2018: 4. -2. 0. 2. 4. 6. 8. 10. 12. 14. 0 200 400 600 800 1000. -
Domain-independent User Satisfaction Reward Estimation forDialogue…
mi.eng.cam.ac.uk/~sjy/papers/ubcm17.pdf20 Feb 2018: As training data, the LEGO corpus [28] is used which con-sists of 200 dialogues (4,885 turns) from the Let’s Go bus infor-mation system [29]. ... Each turnof these 200 dialogues has been annotated with IQ (represent-ing the quality of the dialogue up -
PROBABLISTIC MODELLING OF F0 IN UNVOICED REGIONS IN HMM ...
mi.eng.cam.ac.uk/~sjy/papers/ytgk09.pdf20 Feb 2018: 100 200 300 400 500 600 700 8000. 50. 100. 150. ... 200. 250. 300. Time index (ms) of "And then". F0. extr. -
acl2010.dvi
mi.eng.cam.ac.uk/~sjy/papers/gjkm10.pdf20 Feb 2018: 2. 200 400 600 800 1000 1200 1400 1600 1800 2000 2200 2400 2600 2800 3000.
Search history
Recently clicked results
Recently clicked results
Your click history is empty.
Recent searches
Recent searches
Your search history is empty.