Hide Notes 

Slide 1: WHITE PAPER Key Selection Criteria for Computer Telephony and Speech Application Development Tool 200 - 750 Palladium Dr, Ottawa, Ontario, K2V 1C7, Canada Tel +1.613.271.8989 Fax +1.613.271.8388 www.pronexus.com
Slide 2: WHITE PAPER Table of Contents Programming Environment Ease of Use Industry Standard Language Debugging Architecture Modularity Rapid Application Development (RAD) Tools Speech Processing Automatic Speech Recognition Text-to-Speech Deployment Considerations Support for Standards Application Examples VBVoice Summary 3 3 4 4 4 4 4 5 6 2 © 2010 Pronexus Inc.
Slide 3: WHITE PAPER Even in the current economic climate, the market for computer telephony and speech applications is growing, as organizations are looking for ways to lower costs, improve customer service and optimize operational efficiency. Speech applications, in particular, are finding increased market acceptance. Developers looking to enter the voice business market face a number of hardware and software choices. Chief among these decisions is the choice of development environment – it can determine success or failure of a development initiative and make the difference between a profitable project rollout and a costly failure. This document highlights some of the critical issues that should be considered: • • • • • • • • • • • Ease of use Industry-standard language Debugging Architecture Modularity Rapid application development tools Speech processing Automatic speech recognition Text-to-speech Deployment considerations Support for standards Ease of Use The ideal programming environment should accomplish a number of goals, the most important being ease of use. To maximize ease of use, most telephony and speech toolkits offer a degree of visual programming through a ‘drag-and-drop’ style interface. However, while visual programming has the potential to greatly enhance ease of use, it can also become a limiting factor. A short learning curve sometimes comes at the expense of other developer productivity aspects that may only become obvious after the tool is in use. Extensibility is key among these. The developer’s ability to extend the visual programming environment to 3 Incorporate other components not provided by the toolkit or to take full advantage of new hardware and software is critical for most real-world telephony applications. For example, the ability to integrate with .NET environments and applications is an increasingly significant decision criterion.
Slide 4: WHITE PAPER Industry-standard Language To further shorten the learning curve for developers new to telephony and speech, the programming language should, ideally, be industry-standard (such as C#, VB.NET) and not proprietary to the selected tool. While most development environments for computer telephony applications allow for the addition of external functionality, some require knowledge of proprietary scripting languages or handling of complex, low-level programming. Preferably the application development environment will enable developers to leverage their programming knowledge and expand on the telephony controls included in the toolkit through custom code, as well as make use of third-party components. on Windows®, don't miss out on the benefits of the nextgeneration technology from Microsoft® - your tool must support .NET! Modularity The ability to break your speech application into cooperating modules is a must. Not only does it improve scalability, reliability and performance of your system, but it also saves you money in both development and production. A modular system is cheaper to build and maintain. In development, programmers benefit from working in parallel on well-defined modules. In production, independent module provisioning and software hot-swaps eliminate costly system downtimes. At the same time, separating application logic from telephony and speech processing allows resource sharing, which in turn leads to more efficient utilization. Finally, distributing your modules across a Local Area Network (LAN) enables load balancing and effortless scalability - again resulting in savings on system maintenance. The biggest benefit, however, comes from increased reliability of a modular system. Nothing is more frustrating to callers than a system that crashes into "dead silence" in the middle of a transaction. An unreliable system will be soon pulled out of production, which always means significant financial losses. A monolithic executable is only as reliable as its weakest component, while a modular system can stay operational even after losing one of its modules. Therefore, it is very 4 Debugging The strength of the debugging environment, and the ability of source level debugging is of critical importance. The right development and debugging tools can save your project. Attempting to uncover call flow or recognition problems in an application without a rich set of tools sharply reduces developer productivity and increases error rates of the final application. Ideally, the chosen programming environment should: • • Tie into an industry-standard debugging platform that you or your developers are already familiar with Have the ability to generate sophisticated call log files for further analysis. Together these characteristics allow developers to create sophisticated and powerful applications without a long learning curve and ongoing trial-and-error during application development. If you're building an application
Slide 5: WHITE PAPER important that application modules execute properly separated from each other and from the system processes, so that a fatal error in one doesn't bring down the whole system. The modules should run out-of-process, or even better, distributed across a LAN. Ideally, modules should be compiled directly into stand-alone executables, not into intermediate scripts or p-code. Not only does this speed up program execution, it also removes the dependency on a shared runtime engine as a single point of failure. should ensure that the platform does not create a monolithic executable for multiple applications. In such a system, the failure of one module could stop the entire system, as the monolithic executable is only as reliable as its weakest module. While your code may be bulletproof, can you guarantee the same for all the components and libraries you have to use? Rapid Application Development Tools Architecture Most telephony and speech applications are businesscritical in nature. They serve as auto attendants, automate routine processes in call center environments, schedule employees and much more. Failure of an application to scale to required levels and to survive hardware problems is not an option. The architecture of your platform should offer proven scalability and the ability to hot-swap applications. It is important that applications can execute independently from each other and from the system processes, can be hot-swapped, easily provisioned and configured. In hosting environments or in situations where multiple applications are being deployed, the platform should enable the sharing of telephony resources (i.e. speech licenses and telephony hardware) across multiple independent applications. Such a distributed architecture also allows individual applications to be interrupted for upgrades or other maintenance without interrupting other applications on the same server. While some environments allow the sharing of telephony hardware and hot-swapping of applications, developers The world of telephony and speech applications is a complex one, with multiple hardware and programming interfaces and a plethora of standards. The resulting learning curve for new developers tends to be very steep. This is where rapid application development tools really shine! Their controls encapsulate and abstract common call processes to simplify application development and shield the developer from hardware specific programming. In evaluating your application development tool, you should look not only at the raw number of these controls, but rather at their depth, customizability and extensibility. Otherwise, you may find that the feature you are looking for is simply not doable in the environment of your choice. The tool should also provide advanced protocols such as ISDN and VoIP and offer a comprehensive lineup of call control features for your particular application. For example, call queuing, agent monitoring, recording and conferencing capabilities are significant in call center applications, while other solutions may require broad fax support, web integration, switch integration via TAPI, etc. 5
Slide 6: WHITE PAPER Automatic Speech Recognition The implementation of an effective automatic speech recognition (ASR) solution can reduce the number of agents, supervisors, trainers and quality assurance specialists that are needed by your business. If a consumer is provided the option of gathering the information he or she needs without accessing an agent, more agents are free to handle calls that cannot be resolved with self-service. Current trends: • • Need to reduce call center operating expenses to improve the company's margins and bottom line Desire to automate as many servicing tasks as possible prior to moving call center activities offshore Need to replace self-service applications that are either reaching their end of life or becoming too expensive to maintain Current generation of self-service Web and voice portal technologies and solutions that can deliver high-value applications not previously available. Text-to-speech When you want to personalize a dialogue with customers, it's almost impossible to record every possible message combination. Greeting customers by their name? Informing customers of why their plane is delayed? Letting customers know the name of the person who signed for the parcel they sent? Process online transactions, bill payments, account balances, inventory, flight information, etc. Text-to-speech (TTS) is a technology that allows you to create a real-time link between text-based content in your database and a customer awaiting an immediate reply. TTS can read any text out loud without knowing the vocabulary, the names, the numbers or codes contained in the text. This technology is mature; it has been validated by market deployments and is already largely used in telephony services provided by carriers and enterprises alike. The development platform should support multiple TTS engines. A key attribute here is again the degree of support for native interfaces to allow for maximum control. For example, relying solely on SAPI as the interface to multiple TTS engines may result in problematic implementation. This is because the generic SAPI interface works differently with different engines. Small variations in timing, buffering schemes and performance, for example, can result in irritating gaps, clicks and delays. The flexibility to mix and match TTS engines through a Media Resource Control Protocol (MRCP) Connector is also of importance, giving those running on non-Microsoft platforms the ability to capitalize on the cost-effective 6 • • The market for speech recognition engines is continually evolving. Consequently, the development platform should support multiple ASR engines, letting you select the appropriate engine for each individual application development effort. Additional speech capabilities to look out for: • • Does the tool support speaker verification capabilities of one or more speech engine vendors? What other areas of speech deployment are being supported? Dynamic grammars, for example, can be used to simplify complex application development.
Slide 7: WHITE PAPER Speech Server and TTS products available in Microsoft’s Office Communications Server (OCS) rather than being constrained by the choices available for their operating system. Pronexus has partnered with Aumtech and certified their MRCP Connector with VBVoice. application should be as painless as possible, yet assure the developer of the licensing integrity of the finished product. Consider whether the platform requires the use of “dongles” for commercial application deployment, or whether the process is only software-enabled. The latter approach allows for easy upgradeability and also has the potential for the re-licensing of developed applications to other customers. Other Speech Capabilities When you select a development tool, ensure that it supports the speech engines or platform that you chose for your speech application, whether Microsoft Speech Server or another speech engine. Also, make sure the development tool provides a flexible programming environment that enables you to customize and extend existing building blocks or incorporate other components not provided by the toolkit to take full advantage of new hardware and software. This capability is critical for most real-world telephony applications. Consideration should also be given to the tool’s ability to record and to dynamically alter voice prompts for your applications. This is particularly important for personalized call processing, depending on caller preferences, payment options etc. In these circumstances, prompts and call flows should be easily manipulated at runtime. Needless to say, the recording and editing of prompts should be an integrated capability. Support for Standards Depending on your organization, standards such as .NET or VoiceXML can also play an important role in your tool selection. In the author’s opinion, however, other considerations as outlined in this document will likely be more important in the day-to-day development work. In addition, standards may impose restrictions on your development effort, since certain new developments and/or specific capabilities that you are looking for have not yet been reflected in a generally slower evolving public standard. VoiceXML, for example, has an acknowledged weakness in the area of call control. For the foreseeable future, other platforms will continue to exist, and some of them are beginning to extend to support standards environments. Programming in VB, VB.NET, C#, which are industrystandard programming languages, eliminates the need to learn proprietary languages and shortens the learning curve for developers new to the Interactive Voice Response (IVR) & telephony landscape without the need to worry about the telephony hardware APIs or media 7 Deployment Considerations Another important consideration for developers is the ease of product licensing and deployment. This is particularly true for commercial application developers, such as system integrators and Independent Software Vendors (ISVs). Ideally, deployment of a finished
Slide 8: WHITE PAPER processing layers. Thereby allowing you to rapidly create powerful IVR and voice-enabled communication solutions while significantly reducing your time to market. Why Choose VBVoice? • Create sophisticated IVR applications using familiar programming skills and industry-standard programming languages (VB6, C# and VB.NET) • Leverage leading speech and telephony technologies including ASR, HMP, MRCP, SIP, TTS and VoIP Cut development time with: • • • • • • Visual call flow environment Fully customizable voice controls Event-driven framework Source-level debugging Prompt library (English (UK & US), French, German, Italian, Japanese, Spanish (Castilian & South American)) Other Speech Capabilities While there are hundreds of application types that are possible with an IVR development tool, the following is a list of some of the most common telephony applications: • • • • • • • • • • • Auto-attendant Automated payments Fax applications Gift card activation Market research and polling Patient notification Predictive dialer Self-service Virtual PBX Voice verification VoIP IVR, etc. • Give your project a head start with: Tens of sample applications; e.g. help desk, speech attendant, fax service, predictive dialer, etc. Best of all, VBVoice is now available free of charge! Request your free copy of VBVoice 7 from our website. • VBVoice Thousands of companies all over the world use VBVoice in numerous industries from banking to government, healthcare, insurance and utilities. Service provider and in-house developers can reduce IVR application development time and bring their IVR solutions to market faster with VBVoice. Other VBVoice benefits include streamlining internal processes, being able to provide better customer care and increasing revenues. Summary The choice of a development tool for your telephony and speech application is subject to a number of criteria, each of which requires careful consideration. The ideal tool should offer a satisfactory answer in all of the outlined categories and excel in the ones that count in your particular scenario. Established in 1994, Pronexus specializes in telephony, speech and database integration technologies. Our flagship product VBVoice IVR software enables developing feature-rich inbound and outbound IVR solutions that improve customer service and streamline internal processes. Thousands of companies all over the world use VBVoice in applications varying from auto-attendants to automated payments, fax applications, notifications, polls and surveys. Part of our business model is to provide customers flexibility to choose between developing an IVR application in house, leveraging Pronexus’ Professional Services or buying a turn-key IVR application from one of our partners who have integrated VBVoice into their solutions. 200 - 750 Palladium Dr, Ottawa, Ontario, K2V 1C7, Canada Tel +1.613.271.8989 Fax +1.613.271.8388 www.pronexus.com