I am a Language and Speech technologies Consultant with a Post-Doc in Spoken Dialogue Management, a PhD in Automatic Text Summarisation, an M.Sc. in Machine Translation, and a B.A. in English and Linguistics. I am the founder and Director of DialogCONNECTION Ltd, a Voice User Interface Design (VUI Design) consultancy based in Manchester (UK) and specialise in user-centred and adaptive strategies for mixed-initiative dialogues focusing on spontaneous speech, but also persona design, usability testing and application development, tuning and optimisation for the English, German, Greek and French languages.
I have been working in Natural Language Processing / Language Engineering since 1991 (for over 20 years), initially in the field of Machine Translation, then Automatic Text Summarisation, and since 1996 in ASR / Speech and Voice Recognition. As a result, I am an expert in both written language / text processing and speech processing. Throughout this time I have focused on interactive / dialogue systems, i.e. intelligent (or seemingly intelligent) conversational agents that can mimic both language understanding and language production, but also user modelling, goal modelling, context modelling, and discourse modelling. All of these types of modelling and processing give rise to intelligent, goal-oriented, effective, sympathetic, user-centric and affective behaviours. I have also been following basic and applied research on affective modelling, emotion modelling, and prosodic modelling and how it contributes to the modelling of the interaction and the planning of the next action / utterance / step.
I have spent the last 16 years envisioning, specifying, designing, documenting, developing, testing, tuning, optimising and explaining speech recognition solutions and Voice User Interfaces (VUIs) both independently through my own consultancy, DialogCONNECTION Ltd, and as part of Pre-Sales and Professional Services teams at voice companies. I have been focusing particularly on intuitive, usable, but also robust and efficient applications with high user acceptance. I am extremely knowledgeable in the voice recognition / speech recognition market, as I have worked for dozens of different organisations and voice application providers across dozens of verticals worldwide; including Apple, Raytheon / BBN, Vodafone, Convergys / Intervoice, CapGemini, Cable & Wireless, Barclays Bank, and the UK National Health Service. I have held positions as Senior VUI designer in Pre-Sales and Professional Services teams of several players in the Speech Industry; as a Speech Technology Research Fellow and Technology Advisor for Management Consulting firms, Conferences and trade publications; as a Trainer, Project Manager and Speech Product Development Team Leader. I have seen the voice recognition / speech recognition market grow exponentially in the last 16 years; I have been observing minor to major shifts in focus, as well as new and emerging industry and market trends in the technology but also in the application space, and I have seen related software companies come and go, or come and get M&Aed;! .(I have been working mainly with Nuance, including what was previously Scansoft, Speechworks and Loquendo).
As a result, I can very reliably comment on the competitive landscape (major and top players, monopolies / oligopolies, USPs and winning criteria for the decision-makers). Having examined for decades end-user attitudes towards and experience with voice recognition applications, VUIs and Interactive Voice Response Systems (IVRs), I have become a worldwide expert in identifying, specifying and recommending what will work with a user and what will not (feasibility, technology milestones, service attractiveness, user expectations and preferences). As a result, I have a good overview and insider knowledge of the technological, logistical and commercial issues involved in the corresponding voice and touchtone applications as well as of the user interface and conversational agent market and industry as a whole. I have also an excellent understanding of the voice dictation and voice activation technologies and applications (including medical and legal dictation software and mobile applications, such as those promoted by Apple’s SIRI and Google Voice) and I can comment on their usability, their limitations and their outlook, although I can only comment on individual products to a limited extent, and prefer to focus on technological approaches. I have also been closely following various other fields of Artificial Intelligence and User Interface Design in the past two decades, and can therefore comment on current and future trends in eye tracking, gesture recognition, emotion recognition, affective computing and chatbot applications for instance.
I have (co)authored several publications, including the Chapter on Spoken Dialogue Systems in the Oxford University Press “Handbook of Computational Linguistics”. I have also served in several Advisory Committees of research projects, international conferences and scientific journals; and have been contributing to SpeechTEK with presentations on VUI Design issues. Originally from Greece, I have spent the last two decades between the UK and Germany, and speak 4 languages fluently.