How can chatbots learn from existing VUI design? What makes these new interfaces different, and how are they similar? This conversation discusses text-based vs. voice-based conversation, analytics for bots and voice, and how bots and IVRs might learn from each other.

The new W3C Voice Interaction Community Group is collecting use cases for voice interaction with virtual assistants. We are exploring topics such as discovery of virtual assistants with specific expertise, standard formats for statistical language models, interoperability for conversational interfaces, and work on dialogue management or “workflow” languages. This session introduces the Voice Interaction Community group and solicits your ideas for future topics.

Today it’s possible to build a software application using services from dozens of online resources: databases, media storage, computing power, error monitoring, customer relations, payment services, and far more. Now that we’ve seen the emergence of speech as a service on our cellphones, how close are we to simple, modularized, network-based speech services for recognition, text to speech, identification, verification, and surveys? And what other services would we like to have for our industry?

A panel of longtime speech technology specialists delivers contrasting views on the role of speech technology in the world of conversational commerce. This highly popular panel starts with the question, “Are you long on speech” as a preferred user interface, then discusses where, how, and why it will coexist, augment, or enhance a plethora of smart user interfaces for digital commerce and self-service.

Chatbots have been hailed as an important new human/machine interface that will replace apps and traditional IVR systems. What is new about today’s chatbots? How are they different from previous chatbots such as 50-year-old Eliza, from VoiceXML dialogue systems, and from embodied conversational agents? Is there a danger that developers of today’s chatbots will ignore lessons learned from the past and are in danger of reinventing the wheel unnecessarily?

Natural language (NL) democratization is happening at a dizzying pace with the advent of text bots (chatbots or virtual agents). Most messaging bots cannot deliver a high level of NL accuracy that is table stakes for enterprise IVRs. Learn how a unified technology stack can leverage the same NL models for both chat bots and IVRs and meet and exceed enterprise NL accuracy expectations. Emphasis will be on statistical language models, statistical semantic interpreter, and predictive technology that further enhances NL intent accuracy for both IVRs and bots.

Does it seem like all the businesses around you are hurtling toward digital transformation at warp speed, while you’re still trying to figure out where to begin? Don’t worry; you’re not alone! This presentation provided fresh insights into a more holistic (and more human) approach to digital transformation—an approach that has the potential to change your customers’ lives, not just your technologies.

Conversational commerce now spans apps, virtual agents, and services offered through smartphones, home electronics, automobiles, public kiosks, and elsewhere. This talk describes real-world use cases that integrate speech processing with natural language understanding, analytics, knowledge management, and other enterprise infrastructure. Hear how conversational commerce is taking shape. Learn how the world of apps, virtual agents, and smartphone services relates to each other. Will they compete or cooperate? Will they integrate into some new powerful capability?

The speech experience in the car is transforming from basic command and control to natural interactions with automated assistants. This presentation explores the current embedded speech experience in the car, both inside and outside of the cloud. We then consider cloud-only solutions. Finally, we discuss the optimum speech experience for the driver, and what’s required to achieve this optimum experience. These implications may apply to other mobile devices and the IoT.

Learn how healthcare organizations and research institutions (including a university college of nursing, cancer transplant center, and primary children’s hospital) leverage digital and telecare platforms to empower patients to adhere to a medication regimen, provide adherence visibility to caregivers, and support messaging and visual “telecare” services. The goals are to improve patient satisfaction while reducing costs, improve population health by tracking medication adherence, and support healthy behavior through “voice with visuals” coaching.

Identifor is an innovative learning and assessment website and mobile application for individuals with autism at every level. It uses interactive online gamification and conversational interaction to identify unique abilities, skills, and interests of students. The games are fun and entertaining with results “translated” into core competencies and then made understandable to parents and educators. Identifor helps more than 2,000 students, parents, and educators. A new Life Coach companion app for special needs adults is under development.

Hardware companies face challenges when implementing speechbased services into their products. Challenges include hardware design considerations such as microphone placement, acoustic dampening materials, and digital signal processing components, as well as considerations around trigger word implementation, branding, and the use of multiple voice services. This presentation explores the trade-offs and challenges when adding voice interaction to hardware products.

The world is increasingly filled with devices that can move around and do things on their own. These include toys, robots, drones, vacuum cleaners, lawnmowers, and many other devices. Consider spoken language interaction with devices that move and how this interaction is different from interaction with stationary devices. We discuss safety, real-time requirements, and feedback from a device that could be out of sight, as well as programming and scheduling the actions of an autonomous device.

Does it seem like all the businesses around you are hurtling toward digital transformation at warp speed, while you’re still trying to figure out where to begin? Don’t worry; you’re not alone! This presentation provided fresh insights into a more holistic (and more human) approach to digital transformation—an approach that has the potential to change your customers’ lives, not just your technologies.

Many speech-enabled products are incapable of recognizing users, allowing kids, friends, and complete strangers to control these devices. By fusing speech recognition technology and voice biometrics, only enrolled users can interact with devices and services, subject to restrictions set by the owners. Hear about new deep learning speech recognition technologies that fuse accuracy and performance with voice biometric security technology that will secure users from identity theft, and enable parental controls as the voice revolution continues to take shape.

“Social robotics” has been cast about as a developing field for a couple years, but what happens when an affordable, consumer robot hits the marketplace? This talk looks at the early metrics gathered from the first few months of Jibo’s experience in the field, focusing on why people buy him, what they do with him, and what they hope to see in the future.

Voice-controlled wearables provide retail employees a 360-degree operational awareness and access to the right information at the right time to maximize customer service. By using intuitive structured speech and a custom enterprise wearable, store associates are able to quickly communicate with one another and access corporate business systems to obtain needed information (inventory status, request assistance, product details, sales information) to make the sale and provide great customer service.

IVR systems are deployed widely, with many regular users familiar with the IVR workflow. Learn how to support both IVR workflow dialogs and complex speech utterances at the same time, and then select the correct flow. A natural language generation module, trained using deep learning methods, is used to generate the dialogue responses. Learn how tools and techniques solve the problem of 511 traffic information service.

To the surprise of many technologists, verbal cooperation flouts the rules of math and formal logic—the game of conversation is played with a different set of rules. Come find out what they are, and rethink the rules of grammar design and error repair, how to help users know what to say, how to accommodate users’ diverse communication styles, and how to improve the perception of your application and brand in the minds of everyday people who talk and listen to machines.

When interacting with a virtual assistant, the centerpiece of the user’s experience is the conversation itself. This means that each “error” is an opportunity for the designer to forge a meaningful exchange between the virtual assistant and user. Let’s leverage users’ mental models of how everyday conversations unfold in the negotiation of meaning. Learn how to frame a new way of approaching conversation design, in which so-called errors become organic turns in the dialogue—moving conversational design forward naturally.

Does it seem like all the businesses around you are hurtling toward digital transformation at warp speed, while you’re still trying to figure out where to begin? Don’t worry; you’re not alone! This presentation provided fresh insights into a more holistic (and more human) approach to digital transformation—an approach that has the potential to change your customers’ lives, not just your technologies.

Despite state-issued bans, the use of cellphones while driving is on the rise. How can we manage proper access to smartphone services while driving? Learn how smart, context-aware digital assistants with co-pilot expertise estimate the level of driving risk; evaluate driver attention and ability to respond to emergencies; interact with the driver, knowing when to speak or not; and inform and coach drivers to minimize risks and improve their performance.

Calling a company and dealing with a system-initiated IVR is the extent of most people’s experience interacting with a spoken dialogue system. This will change with social robots. What is the etiquette when conversing with a robot? Who talks when? What are the boundaries of acceptable things to say? With each evolutionary step forward, the rules of turn-taking shift. We focus on the fascinating challenges faced by the Jibo design team while creating the foundations for a human-robot conversation.

The RBC Conversational Customer Care Virtual Assistant was designed to boost automation, minimize advisor-to-advisor transfers, improve customer satisfaction with conversational dialogues, and use voice biometrics for user authentication. Taking a holistic approach for conversational customer care, everything is deployed on a single platform with a single knowledgebase, able to work over all available channels, and able to connect to real human agents when necessary to complete the goal of servicing 100% of customer requests on the first interaction.

Utility customers used our old IVR system to report outages, pay bills, hear balances, provide meter readings, start or stop services, and more. To self-serve as many callers as possible and intelligently transfer other callers, we replaced the existing directed dialogue with NLU. Learn how we uncovered and resolved problems including utterances that were bucketed incorrectly, callers who were being reprompted at the SLM too many times, and disambiguation questions that weren’t clear.

Statistical language model (SLM) grammars have been shown to have a positive effect on customer experience. However, there is often a lack of appreciation of complexities surrounding the building, deployment, and support of SML. This session examines different scenarios and suggests criteria to help gauge whether an application needs an SLM-type solution or not. Additionally, a comparison is presented of the performance of SLM grammars vs. rule-based grammars on some of the most challenging recognition tasks.

If your company is looking to implement natural language, what metrics are you going to use for determining success? This talk will compare several natural language deployments to help you understand when NL is a good fit and the metrics you should consider tracking to determine success. We’ll compare and contrast several NL deployments across verticals to understand how those industries created their business case and the deployment results.

Does it seem like all the businesses around you are hurtling toward digital transformation at warp speed, while you’re still trying to figure out where to begin? Don’t worry; you’re not alone! This presentation provided fresh insights into a more holistic (and more human) approach to digital transformation—an approach that has the potential to change your customers’ lives, not just your technologies.

Teams that use rule-based approaches to build demos of voice or chat assistants run into trouble when trying to take those apps to production. This happens because as you go beyond “toy” demo functionality, the rules rapidly become unworkably complex. To handle human language in all its endless variation, statistical NLU models are needed. This talk shows how you can build production-ready conversational apps in ten steps, using machine-learned language models and AI algorithms.

451 Research survey data shows that consumers are increasingly adopting connected devices and applications, such as Amazon’s Echo and virtual assistants such as Cortana and Siri. We expect these trends to extend into the enterprise. This session looks at specific use cases that represent near-term opportunities where speech technologies transform the user experience and redefine how organizations create and deliver value. Key challenges for adoption of speech-enabled devices and applications in the enterprise are identified.

The goal of our solution, Visual IVR, is to transform voice IVRs into a personalized and digital experience, improving self-service by more than 10% based on our implementations at Fortune 500 companies globally. Visual IVR also allows customers to transition from the digital channel to the agent with full context resulting in significantly reduced handle times. Visual IVR does not need an installed native mobile app—ensuring higher customer adoption.

Rogers has started down a omni-channel customer service path by using digital channel customer interactions to “predict” why the customers are calling and to offer targeted messages, self-service options, and routing shortcuts in the Rogers IVR. We describe the wins, lessons learned, and next steps of this multi-year journey.

Mix and mingle with other attendees as well as speakers and our conference sponsors during this networking reception. Admission is included with the purchase of a Discovery Pass or higher. (Not included with an Exhibits-Only pass.)