NPFL123 Dialogue Systems 1. Introduction
NPFL123 Dialogue Systems1. Introductionhttps://ufal.cz/npfl123Ondřej Dušek, Vojtěch Hudeček & Jan Cuřín2. 3. 2021Charles UniversityFaculty of Mathematics and PhysicsInstitute of Formal and Applied Linguisticsunless otherwise stated
Organizational: NPFL123 – 2/2 C Ex – 5 Credits Lecture (Tue 10:40am) labs (Tue 9am/9:50am) Lecture: intro, theory Labs: practical examples, hands-on exercises To pass the course: Written exam – 10 freeform questions, covered by the lectures Lab exercises – weekly small homework implementing your system for a domain other small dialogue-related exercises Slides, news etc. at ufal.cz/npfl123NPFL123 L1 20212
About Us Ondřej Dušek: lectures, course guarantor PhD at ÚFAL, 2 years at Heriot-Watt Uni Edinburgh, now back worked mostly on language generation also chatbots (HWU Alexa Prize team) Vojtěch Hudeček: some labs, a bit of lectures PhD student at ÚFAL (4th year) working on dialogue management & language understanding internships at Uber AI & UC Davis on dialogue systems Jan Cuřín: dialog authoring tools, speech recognition The MAMA AI – Co-founder, 16 years at IBM Prague Research Lab PhD at ÚFAL in 2006 (machine translation) Dialog systems and applications, speech recognition, machine translationNPFL123 L1 20213
Course Syllabus1. Introduction (today)2. What happens in a dialogue?3.4.5.6.7.8.9.10.11.12.Dialogue system data & how to evaluateAssistants (Alexa, Siri, Google etc.), question answeringDialogue authoring/tooling systemsLanguage understanding (NLU)NLU Dialogue state trackingDialogue management (DM)DM Language generationSpeech recognitionSpeech synthesisChatbotsNPFL123 L1 20214
Recommended Reading There’s nothing ideal (active research topic!)Primary (brief): Jurafsky & Martin: Speech & Language processing. 3rd ed. draft 2020, Chap. 24(https://web.stanford.edu/ jurafsky/slp3/)Other (see also website): Gao et al.: Neural Approaches to Conversational AI, 2019 (http://arxiv.org/abs/1809.08267) McTear et al.: The Conversational Interface: Talking to Smart Devices. Springer 2016. Janarthanam: Hands-On Chatbots and Conversational UI Development. Packt 2017. Skantze: Error Handling in Spoken Dialogue Systems. PhD Thesis 2007, Chap. 2(http://www.speech.kth.se/ gabriel/thesis/chapter2.pdf) Jokinen & McTear: Spoken dialogue systems. Morgan & Claypool 2010. Psutka et al.: Mluvíme s počítačem česky. Academia 2006. Lemon & Pietquin: Data-Driven Methods for Adaptive Spoken Dialogue Systems. Springer 2012. Rieser & Lemon: Reinforcement learning for adaptive dialogue systems. Springer 2011.NPFL123 L1 20215
What’s a dialogue system?Definition: A (spoken) dialogue system is a computer system designed to interactwith users in (spoken) natural language Wide definition – covers lots of different casesNPFL123 L1 20216
“AI”: sci-fi vs. realityhttps://youtu.be/1ZXugicgn6U?t 3 Lots of talk about AI now Hype around Siri/Alexa/Google Sci-fi expectations – AI-complete Star Trek – know-it-all 2001 Space Odyssey –mutiny Her – personalityhttps://youtu.be/qDrDUmuUBTo We’re not there – probably for long main bottleneck: understanding(not speech comprehension, meaning!) more like the Red Dwarf talkie toasterNPFL123 L1 2021https://youtu.be/6QRvTv tpw0?t 27https://youtu.be/LRq SAuQDec?t 717
Example: Google Assistant Handling call for a client (Google IO 2018 demo) very natural speech show’s what’s possible now in a limited domain redirects to a human if it can’t handle a shop’s requesthttps://youtu.be/d40jgFZ5hXkNPFL123 L1 20218
Example: Alana Chatbot (Heriot-Watt University) Open-domainNPFL123 L1 ://www.dailymotion.com/video/x7takq4(the Alana section starts at 47:38)9
Why take interest in Dialogue Systems? I
4. Assistants (Alexa, Siri, Google etc.), question answering 5. Dialogue authoring/tooling systems 6. Language understanding (NLU) 7. NLU Dialogue state tracking 8. Dialogue management (DM) 9. DM Language generation 10. Speech recogn
approach and a dialogue editor to write these dialogue scripts. Online chat bots are proposed as a technique to evaluate and to improve an interactive dialogue script. Results of a pilot study with 4 non-phobic individuals are promising and suggest that these scripted interactive dialogues can be used to simulate a human-human dialogue.
This report describes the state of the art of the dialogue management research in a context of both spoken and multimodal dialogue systems. Section 2 describes an overview of a multimodal dialogue .
ministere du travail, de l’emploi et des sportset du dialogue social ministere de la ville, de la jeunesse rapport de synthese des travaux du groupe de dialogue inter - partenaires sur la lutte contre les discriminations en entreprise président du groupe de dialogue: jean-christophe sciberras
dialogue act model, which relies on hierarchical clustering and uses no manual labels during model training, would support equal or better student learning than the supervised dialogue act model, which relies on a decision tree classi- er and represents a state-of-the-art, highly accurate dialogue act classi er that
Apr 30, 2016 · ‘Difficult Dialogue’ resources. These are composed of two discrete sections. The first is this book – Difficult Dialogue in the Classroom, which builds upon the work outlined in the Essentials of Dialogue, to outline an approach to addressing such issues using tried and tested
WORKSHEETS MIND YOUR LANGUAGE ASKING RESPONSE QUESTIONS SKILLS OF DIALOGUE OTHER SKILLS DIALOGUE CHECKLIST PRACTISE MAKES PERFECT ACTIVE LISTENING . Active Listening is not just paying attention but is a specific methodology that enables practitioners to deepen dialogue to improve comm
83. Ecumenical Dialogue in the Orthodox Coptic Church Wedak Tawfi k 525 84. Ecumenical Dialogue in the Ethiopian Orthodox Tewahedo Church Ayalkibet Berhanu 534 85. Ecumenical Dialogue in the Syrian Orthodox Church Archbishop Philoxenos Mattias Nayis 546 86. Ecumenical Dialogue in the Malankara Orthodox Syrian Church Kondothra M. George 552 87.
aliments contenant un additif alimentaire des dispositions des alinéas a) et d) du paragraphe 4(1) ainsi que du paragraphe 6(1) de la Loi sur les aliments et drogues de même que, s'il y a lieu, des articles B.01.042, B.01.043 et B.16.007 du Règlement sur les aliments et drogues uniquement en ce qui a trait
