Before you make changes to your IVR, it’d be nice to get a guarantee that the savings from the changes will cover the investment costs. This presentation describes a method of using analytics to get such a guarantee.

When enterprises undertake speech technology projects in piecemeal fashion, one point solution at a time, the result is application silos, fragmented architectures, and disjointed caller experiences. Today, callers expect integrated service, and with the right technology and methodology, an enterprise can go beyond point solutions to automate a much higher proportion of caller minutes. This presentation discusses how enterprises are rolling out integrated suites of speech applications, fronted by intelligent call steering, that share personas, service processes, caller data, business rules, and backend integration.

In the new world of open, mobile communications, oftentimes, the user’s location is unknown. This presentation covers some of the new carrier network location capabilities and geospatial call routing. Speech and text can be combined with location to deliver a richer, more relevant user experience. In addition, using multimedia messaging, phone callers can receive maps, pictures, and videos that enhance the user experience.

Many mobile devices now allow complex combinations of input and output modalities, so multimodal UI developers now have a greater responsibility to design interfaces that feel natural to people in spite of higher  complexity. AT&T Research shows that successful multimodal systems account for the communicative process of “grounding,” through which users incrementally provide evidence of the context of an interaction. This talk discusses the design challenges to learning how users’ spoken interactions, spatial movements, and touch gestures on a map contributed to our understanding of grounding’s critical role in the success of a multimodal search interface.

Are customers concerned when an enterprise shares information about their recent behavior as they cross channels of interaction? How much information sharing is enough, and how much is too much? This session presents the results of a study that reveals the perspectives of both real customers and contact center representatives. The results will surprise you! Plus, the presenter shares usability best practices regarding cross-channel interactions.

While automated phone support remains an important link in customer conversations, a recent survey, conducted by Vocalabs on behalf of Nuance Communications, showed that enterprises need to implement a multichannel approach (through IVR, chat, social media, email, and web self-service) to keep up with the ever-changing needs and requirements of their customers.

With the explosion of smartphones and customized enterprise applications, the population that remains loyal to the IVR has shifted. This talk outlines how today’s typical IVR customer differs from customers a decade ago and uses data from deployments to support the argument that the changes in caller demographics result in different usage patterns and customer expectations. Finally, we make the case that these demographic changes challenge us to revisit and recalibrate some of our tried-and-true design best practices for IVR applications.

The persona of a speech application is as critical for a new caller’s first impression as it is to a power user’s repeated success. Persona development has traditionally been a qualitative exercise led by the judgments of speech experts and the subjective opinions of a few business stakeholders. This talk presents a detailed scientific methodology to select the best voice actor to convey the desired persona.

This talk presents high-level initial findings of a comprehensive study of caller behavior at a prompt-by-prompt level to determine which wording works the best. The study includes millions of calls across many vertical  domains. It focuses on many of the most common types of prompting, such as date of birth, credit card numbers, and language preference, and reveals the results of the best-performing wording for each.

As VUI designers, we strive to create a natural experience for our users that takes into account external distractions, the ephemeral nature of speech, and the unnaturalness of talking to a machine optimized for a small set of tasks.  This talk covers the mechanisms of human conversation that enable efficient and informative interaction and their adaptation to dialogues with a minimally sentient nonhuman partner.

Name recognition and synthesis can dramatically improve customer experience, disambiguate confusion, and make interactions more clear and personalized. Because they do not follow the same pronunciation rules as words, names are notoriously difficult to implement well. This presentation shows you how to avoid common pitfalls in name recognition and synthesis, discusses the problems associated with obtaining good data for name pronunciation, and describes best practices for developing good software models.

This talk discusses the behavior of two distinct user groups interacting with automated speech systems for collecting addresses. One group of callers was hyper-careful, speaking very slowly, with long pauses between words, often spelling out words and giving full ZIP+4 numbers. The second caller group was much less careful, spoke more quickly, and there was often side speech and background noise during the call. This presentation compares these two groups and describes how Novauris adjusted the design to accommodate their behavior.

This session describes industry best practices for using speech analytics in the contact center; outlines the trend toward real-time data analysis and howit provides a competitive edge; describes the path toward improving key metrics, including first call resolution, average handle time, and average speed toanswer; and details specific steps to positively impact agent productivity and quality assurance without adding additional staff.

Teams suffer from a lack of usable data to determine design improvements. Neither of the two common solutions — requesting custom instrumentation, or spending hours in manual analysis — is sustainable. This session examines alternative approaches to help analysts uncover application improvement opportunities and discusses the outcomes of prototypes used to test these approaches. The presentation concludes with a recommended process to automatically uncover the largest-impact opportunities to increase application performance, based on actual user behavior.

Learn how to leverage the strengths of each channel for better interactions across and between channels. Companies not only tend to see channels in silos, but they organize them that way too. This session discusses the strengths of different channels and how to leverage them, as well as how to understand the value of delivering service where channels intersect. Examples include SMS surveys after a phone interaction, high-tech/high-touch approaches, and blending channels to deliver service in the best medium possible (speech, visual, etc.).

Providing multichannel communication is critical to enhancing customer satisfaction and improving retention and loyalty. This session defines the core elements needed to create an effective on-demand multichannel communication strategy for success, including agents and automation, two-way voice self-service, SMS messaging, and reporting requirements.

Mobile applications give your company the ability to provide your customers with a rich interaction experience by providing instant access to information about your company. All of the good feelings these applications promote are lost when your customer calls the contact center and has to restart his or her interaction with your company from scratch. This presentation talks about how to create a seamless customer experience that allows you to use mobile, network speech, and access to the contact center in a way that feels like a single transaction.

Customers no longer accept disconnected, channel-specific applications to handle their need for service. This session focuses on how the enterprise can manage service channel progression by using data “interceptors” and speech together to meet the opportunities presented by these new customer expectations. It explores the types of applications best suited for service interceptors, key cross-channel considerations for resuming service effectively, and how speech can be used as an integral tool in any service channel progression strategy.

While speech technologies are widely used in many call/contact centers, video technology is rarely utilized today. With the increase in adoption of IPbased technologies in the contact center and the roll-out of 3G wireless networks, video is increasingly seen as a viable technology. This session discusses how organizations can combine video with the power of speech to expand, enrich, and improve the effectiveness with which they interact with their customers.

Avaya has enabled a PC or mobile device to make two-way voice and video calls to contact center speech systems and live agents through the use of Adobe Flash technology and SIP technology — and then support immediate desktop collaboration between agents and consumers using information provided to these automated voice and video self-service systems. This presentation demonstrates a series of use scenarios, including bank/financial, technical help desk, and a medical tourism/support.

Learn how progressive companies are combining speech analytics solutions with other forms of customer contact such as chat, email, and social media. Attendees learn the details of a customer interaction analytics strategy, where they will be able to merge channels, provide early warning systems from different sources, conduct deep-dive root-cause analysis, and drive information to key stakeholders.

This session looks at one solution to build multichannel applications across voice, SMS, IM, mobile web, and Twitter that results in a common analytics engine across all channels and even allows that data to be brought into common industry business intelligence systems. Learn how you can “develop once and deploy anywhere” to understand how your customers are interacting with your application in whatever communications channel they choose to use.

Learn about various authentication and identity strategies for speech-enabled applications, including speech application design, implementation, and business issues surrounding authentication and identity. Learn about new multifactor, multimodal authentication technologies such as speaker verification and SMS tokens. The session delves into how executives see the growing emphasis on identity and how identity management systems impact speech applications in their organizations today and into the future. A brief overview of the AVIOS Forum Security and Identity committee activities is shared.

Adoption of voice biometric speaker verification is growing, but how secure is it? This presentation describes voice biometric approaches, discusses their resistance to replay attacks, compares capabilities and proposes a secure strategy for automated speaker authentication and real-time identity alerts. Important points and challenges in the state-of-art of voice biometrics are discussed.

To provide a great customer experience, we must determine the reason for a call by integrating the enterprise CRM and backend systems with the IVR. The step to predicting the reason for a call is being proactive, e.g.,using outbound calls to reach customers before they call. Learn how outbound calls can be used to meet the growing customer demands. The session also includes how analytics, user interface, voice biometrics, and business process impact the effectiveness of both inbound and outbound solutions.

Customer relation management agents have to constantly search and retrieve information quickly from structured, semistructured, and unstructured data. This session presents an automatic and powerful spoken-language understanding solution by seamlessly integrating natural language processing strengths with semantic technology capabilities that allow systems to automatically extract, index, and query knowledge from various data  sources. Accurate information is quickly retrieved from the rich body of knowledge to answer spoken-query texts produced by the speech recognizer.