33 Hits in 3.4 sec

Statistical dialog management applied to WFST-based dialog systems

Chiori Hori, Kiyonori Ohtake, Teruhisa Misu, Hideki Kashioka, Satoshi Nakamura
2009 2009 IEEE International Conference on Acoustics, Speech and Signal Processing  
manage dialog reasonably on the WFST-based dialog management platform.  ...  Index Terms-WFST-based dialog system, statistical dialog management, Interchange Format  ...  EVALUATION EXPERIMENTS Evaluation data To validate performance of WFST-based statistical dialog management, we constructed Japanese and English dialog systems for hotel reservations using the corpus.  ... 
doi:10.1109/icassp.2009.4960703 dblp:conf/icassp/HoriOMKN09 fatcat:7rcayjn6ujakzkw6j6teiqr3ru

Efficient integrated response generation from multiple targets using weighted finite state transducers

Ivan Bulyko, Mari Ostendorf
2002 Computer Speech and Language  
In this paper, we describe how language generation and speech synthesis for spoken dialog systems can be efficiently integrated under a weighted finite state transducer architecture.  ...  The choice of wording and prosodic structure are then jointly optimized with unit selection for waveform generation in speech synthesis.  ...  This material is based upon work supported by the National Science Foundation under Grant No. (IIS-9528990).  ... 
doi:10.1016/s0885-2308(02)00023-2 fatcat:uknb4xvrn5g6tdxfckuuo6u6ba

LVCSR System on a Hybrid GPU-CPU Embedded Platform for Real-Time Dialog Applications

Alexei V. Ivanov, Patrick L. Lange, David Suendermann-Oeft
2016 Proceedings of the 17th Annual Meeting of the Special Interest Group on Discourse and Dialogue  
The system is trained on a standard 1000hour corpus, LibriSpeech, features a trigram WFST-based language model, and achieves state-of-the-art recognition accuracy.  ...  The fact that the system is realtime-able and consumes less than 7.5 watts peak makes the system perfectly suitable for fast, but precise, offline spoken dialog applications, such as in robotics, portable  ...  Further advantages of using a low-footprint highly accurate real-time able speech recognizer over cloud-based recognition include • no need for complex load balancing, instance management, or distributed  ... 
doi:10.18653/v1/w16-3627 dblp:conf/sigdial/IvanovLS16 fatcat:pr23zdkicnbtnfgwnhoavbj56q

Editorial: Special Issue on the Eighth Dialog System Technology Challenge

Seokhwan Kim, Hannes Schulz, Chulaka Gunasekara, Chiori Hori, Abhinav Rastogi, Luis D'Haro
2021 IEEE/ACM Transactions on Audio Speech and Language Processing  
She built a WFST-based speech recognition system utilized in commercial products such as spoken dialog systems of NTT DOCOMO and speech translation systems of KDDI AU on smartphones.  ...  On the other hand, the human evaluations are too expensive to make it scalable and relatively less reproducible compared to the conventional corpus-based evaluation methods.  ... 
doi:10.1109/taslp.2021.3097842 fatcat:uyvus3pitrh53ehs5dnwppmile

Web-based environment for user generation of spoken dialog for virtual assistants

Ryota Nishimura, Daisuke Yamamoto, Takahiro Uchiya, Ichi Takumi
2018 EURASIP Journal on Audio, Speech, and Music Processing  
In this paper, a web-based spoken dialog generation environment which enables users to edit dialogs with a video virtual assistant is developed and to also select the 3D motions and tone of voice for the  ...  In our proposed system, "anyone" can "easily" post/edit contents of the dialog for the dialog system.  ...  Acknowledgements This study was supported by the Core Research for Evolutional Science and  ... 
doi:10.1186/s13636-018-0142-8 fatcat:rxoy3pje6zacdarwkxgt2chosa

Using semantic analysis to improve speech recognition performance

Hakan Erdogan, Ruhi Sarikaya, Stanley F. Chen, Yuqing Gao, Michael Picheny
2005 Computer Speech and Language  
In this study, we propose three new language modeling techniques that use semantic analysis for spoken dialog systems.  ...  Although syntactic structure has been used in recent work in language modeling, there has not been much effort in using semantic analysis for language models.  ...  Acknowledgements The authors thank Adwait Ratnaparkhi for the use of his code implementing maximum entropy training and testing algorithms and Mike Monkowski for designing grammars in the financial domain  ... 
doi:10.1016/j.csl.2004.10.002 fatcat:swilwbulpjgqloma5g2lhol43i

A distributed cloud-based dialog system for conversational application development

Vikram Ramanarayanan, David Suendermann-Oeft, Alexei V. Ivanov, Keelan Evanini
2015 Proceedings of the 16th Annual Meeting of the Special Interest Group on Discourse and Dialogue  
This cloud-based spoken dialog system can be accessed both via telephone interfaces as well as through web clients with WebRTC/HTML5 integration, allowing in-browser access to potentially multimodal dialog  ...  We have previously presented HALEFan open-source spoken dialog system-that supports telephonic interfaces and has a distributed architecture.  ...  Acknowledgements The authors would like to thank Lydia Rieck, Elizabeth Bredlau, Katie Vlasov, Eugene Tsuprun, Juliet Marlier, Phallis Vaughter, Nehal Sadek, and Veronika Laughlin for helpful input in  ... 
doi:10.18653/v1/w15-4658 dblp:conf/sigdial/RamanarayananSI15 fatcat:x3obblgwtvf37h3t2zl2k5a6uq

Using probabilistic logic for dialogue strategy selection [chapter]

Ian O'Neill, Philip Hanna, Anbu Yue, Weiru Liu
2011 Proceedings of the Paralinguistic Information and its Integration in Spoken Dialogue Systems Workshop  
. . . . . . . . . . . . . . . . . . . 62 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62 Domain-Adapted Word Segmentation for  ...  . . . . . . 293 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 293 SpeechEval: A Domain-Independent User Simulation Platform for  ...  task-incomplete dialogs . . . . . . . . . . . . . . . . . . . . . . . . . 220 4.1 Evaluation of off-line detection . . . . . . . . . . . . . . . . . . . . . . 220 4.2 Evaluation of on-line detection  ... 
doi:10.1007/978-1-4614-1335-6_25 fatcat:hwdid7te7vdnpeto5xuobu37li

Bayesian Learning of a Language Model from Continuous Speech

Graham NEUBIG, Masato MIMURA, Shinsuke MORI, Tatsuya KAWAHARA
2012 IEICE transactions on information and systems  
Implementation is performed using weighted finite state transducers (WFSTs), which allow for the simple handling of lattice input.  ...  We propose a novel scheme to learn a language model (LM) for automatic speech recognition (ASR) directly from continuous speech.  ...  This is done by first creating a WFST-based formulation of the WS model (Sect. 4.1), then describing a dynamic programming method for sampling over WFSTs (Sect. 4.2).  ... 
doi:10.1587/transinf.e95.d.614 fatcat:eo6m2lm4lfctphf4ri2weurzzq

A parallel meeting diarist

Gerald Friedland, Jike Chong, Adam Janin
2010 Proceedings of the 2010 international workshop on Searching spontaneous conversational speech - SSCS '10  
We therefore developed novel parallel methods for speaker diarization and speech recognition that are optimized to run on multicore and manycore architectures.  ...  The following article presents an application for browsing meeting recordings by speaker, keyword, and pre-defined acoustic events (e.g., laughter), which we call the Meeting Diarist.  ...  [9] compared sequential and parallel implementations of the WFST-based recognition network representations.  ... 
doi:10.1145/1878101.1878114 fatcat:5bam27jtmvf2vdmaobgzyv53eu

Speech Recognition: Statistical Methods [chapter]

L.R. Rabiner, B.-H. Juang
2006 Encyclopedia of Language & Linguistics  
Dialog management systems are evaluated based on the speed and accuracy of attaining a well-defined task goal, such as booking an airline reservation, renting a car, purchasing a stock, or obtaining help  ...  The computational models for dialog management include both structure-based approaches (which models dialog as a predefined state transition network that is followed from an initial goal state to a set  ... 
doi:10.1016/b0-08-044854-2/00907-x fatcat:xmlxpxxcpvcwbl7w3wns47wkna

Miscommunication handling in spoken dialog systems based on error-aware dialog state detection

Chung-Hsien Wu, Ming-Hsiang Su, Wei-Bin Liang
2017 EURASIP Journal on Audio, Speech, and Music Processing  
Historical information-based n-grams are employed to find the most likely DS for the SDS. Several experiments were performed with a dialog corpus for the restaurant reservation task.  ...  This paper presents an approach to error-aware dialog state (DS) detection for robust miscommunication handling in an SDS.  ...  A POMDP-based dialog manager is then employed for dialog management. The remainder of this paper is organized as follows. Section II describes the corpus collection and annotation.  ... 
doi:10.1186/s13636-017-0107-3 fatcat:unvxnnjqdzhyvagbjyr64x7jve

Parallelizing Speaker-Attributed Speech Recognition for Meeting Browsing

Gerald Friedland, Jike Chong, Adam Janin
2010 2010 IEEE International Symposium on Multimedia  
This paper presents the underlying parallel speaker diarization and speech recognition realizations, a comparison of results based on NIST RT07 evaluation data, and a description of the final application  ...  The following article presents an application for browsing meeting recordings by speaker and keyword which we call the Meeting Diarist.  ...  Parallel WFST-based LVCSR is also implemented on CPU and GPU in [18, 4] . [18] compared sequential and parallel implementations of the WFST-based recognition network representations.  ... 
doi:10.1109/ism.2010.26 dblp:conf/ism/FriedlandCJ10 fatcat:odplkgewt5fapn2ew6fkidwbwi

Snips Voice Platform: an embedded Spoken Language Understanding system for private-by-design voice interfaces [article]

Alice Coucke, Alaa Saade, Adrien Ball, Théodore Bluche, Alexandre Caulier, David Leroy, Clément Doumouro, Thibault Gisselbrecht, Francesco Caltagirone, Thibaut Lavril, Maël Primet, Joseph Dureau
2018 arXiv   pre-print
We are indebted to the community of users of the Snips Voice Platform for valuable feedback and contributions.  ...  Once the user's request has been processed and based on the information that has been extracted from the query and fed to the device, a dialog management component is responsible for providing a feedback  ...  Acoustic model evaluation In this section, we present an evaluation of our acoustic model for English.  ... 
arXiv:1805.10190v3 fatcat:ej65i7jecvatlp7wppshiptxwm

Mem2Seq: Effectively Incorporating Knowledge Bases into End-to-End Task-Oriented Dialog Systems [article]

Andrea Madotto and Chien-Sheng Wu and Pascale Fung
2018 arXiv   pre-print
End-to-end task-oriented dialog systems usually suffer from the challenge of incorporating knowledge bases.  ...  As a result, we show that Mem2Seq can be trained faster and attain the state-of-the-art performance on three different task-oriented dialog datasets.  ...  Statisti- cal dialog management applied to wfst-based di- alog systems. In IEEE International Conference on Acoustics, Speech and Signal Processing, 2009. ICASSP 2009., pages 4793-4796. IEEE.  ... 
arXiv:1804.08217v3 fatcat:2wiawhghwzekbfpnbfj5qbgzbe
« Previous Showing results 1 — 15 out of 33 results