Q&A: David Attwater on the Ins and Outs of Conversation Design

Article Featured Image

David Attwater, Senior Scientist, Enterprise Integration Group recently answered the following questions about Conversational Design: 

Q: Tell us about the three-hour workshop you will present on April 26 at the SpeechTEK Conference in Washington, DC.

A: With the explosion of smart-speakers, chatbots, and high-quality cloud speech technologies, conversational interfaces are burgeoning. In spite of the progress of artificial intelligence, most conversational technologies still require a high degree of design skill to make them usable and effective. In this class, we give a very practical deep-dive into the discipline of conversational design. Attendees of this class will get a brief primer into the current technologies, including an understanding of why good design is still so necessary in spite of recent advances in AI. Using group exercises you learn how to select the target tasks for a conversational user interface, and then design a small conversational interface. Topics covered will be task selection, intent design, scripting, and structuring your natural conversation to make it discoverable and stable. This course is open to anyone who wants to understand conversational design better. You could be an engineer, UX designer a strategic manager. The course is appropriate for those considering or working with conversational solutions for call centers, smart speakers, kiosks or chatbots. It is not technical and no prior experience is necessary. 

Q:  Is there a difference in the way you design voice IVRs and other Voice first applications such as Amazon Alexa skills?

A: The principles of conversational design are the same for all spoken user interfaces but this does not mean that the design solutions are the same between IVRs and smart speakers.  The context of the user makes a big difference in the design of the user interface. The most important difference is that with smart speakers the user has chosen to speak with the machine whereas with IVRs they have not.  This changes the initial engagement between the user and the machine. In the class, we will teach the underlying principles and help you understand which principles to apply in which situations.

Q: How advanced is conversational artificial intelligence? 

A:  Speech recognition and text to speech technology have advanced a lot in the last few years.  The accuracy of speech recognition and the naturalness of text to speech are very high. The underlying technologies have been very similar for many years. Computers are faster and can process much more data now and this has led to the recent improvements in conversational AI.  AI is a broad term and it covers two main scenarios: A) Situations where machines can learn patterns from large amounts of data automatically – and B) Situations where designers have carefully crafted rules to simulate intelligent responses. Conversational AI is a mixture of both approaches. To hold a conversation between a machine and a person skillful design is still needed to anticipate the conversational paths and make sure that the conversation is robust and makes sense to the user.

Q: What are the three big take-away messages from this conversational design course?

A: The key messages of the conversational design course are: 

  • Conversational design is a learnable skill with established design principles.   
  • Conversational design and visual user interface design are very different because of the fundamental differences in the channels used.   
  • Understanding conversational design principles is vital to make sure you get the best out of your investment in conversational tools and platforms.

Q:  Should conversational machines behave just like people?

A:  Yes and no. In order to communicate effectively, conversational machines must obey the underlying rules of conversation such as turn-taking, co-operation, and displaying understanding etc. There are very good reasons however why conversational should not pretend to be people and may ask questions or present information slightly differently to the way people do it with each other.  The reasons for these differences will be explained in the class but the primary difference is that conversational machines do not have the deep social understanding and wide world-knowledge that people possess. This means that their capabilities, whilst impressive, are bounded in a way that people are not.

Register for the SpeechTEK Conference and Attwater’s workshop. There are still openings for SpeechTEK University workshops and presentations. Submit proposals here by October 11, 2002.

SpeechTek Covers
for qualified subscribers
Subscribe Now Current Issue Past Issues
Related Articles

Q&A: Dr. Nava Shaked on Evaluation, Testing Methodology & Best Practices for Speech-Based Interaction Systems

Get a sneak-peak into Dr. Nava Shaked's SpeechTEK workshop in this Q&A. Learn everything you need to know about evaluation, testing, and best practices for speech-based interactions.

Q&A: Bruce Balentine on the Basics of Conversational Chatbots

Conversational chatbots may be the future (or even the present) but Bruce Balentine is taking developers back to basics in his workshop at SpeechTEK in April. Learn how to build chatbots that get results.

Q&A: Deborah Dahl on Natural Language Understanding

Jim Larson talked to Dr. Deborah Dahl, Principal, Conversational Technologies about the increasing importance and capabilities of natural language processing, speech recognition, and

How Human Is Too Human?

We hear, quite often, that conversational AI need to be more human. What this means is that we want it to perform for us at the same level as a human assistant, but we still want to know it's a machine.

AI for Speech, Other Uses Primary Message from NICE Interactions

Delivering and leveraging artificial intelligence throughout the contact center, including several AI-enabled speech capabilities was the theme of the NICE Interactions 2019 Conference, which drew an estimated 3,000 attendees.