Voxeo Talks

We are very pleased to announce today that Dan Burnett, Voxeo’s Director of Speech Technologies, has joined the VoiceXML Forum Board of Directors. The VoiceXML Forum issued a news release today announcing Dan’s new role.

Dan Burnett’s passion about VoiceXML and related technologies is no secret to anyone in the industry. He currently serves as the co-editor-in-chief of the VoiceXML 3.0 specification and is actually the chair of the W3C’s Voice Browser Working Group, the group under which speech-related standards fall including VoiceXML, CCXML, SCXML and so many more.

Dan has in fact been involved with many of those standards, serving as:

• co-chair of the HTML Speech Incubator Group
• chief editor of Speech Synthesis Markup Language (SSML) versions 1.0 and 1.1;
• editor of VoiceXML 2.0, 2.1, and 3.0;
• editor of Media Resource Control Protocol(MRCP) v2;
• an author of the Extensible Multi-Modal Annotation (EMMA) specification, Pronunciation Lexicon Specification (PLS), and State Chart XML (SCXML); and
• a contributor to the Speech Recognition Grammar Specification (SRGS), Semantic Interpretation for Speech Recognition(SISR) and the multimodal framework documents of the Multimodal Interaction Working Group.

Basically, if there has been a W3C standard involved with speech or voice, odds are pretty good that Dan has been involved in some form.

Within Voxeo, Dan has overall responsibility for all of Voxeo’s speech recognition products and research. He leads the research, development and implementation of the company’s own speech recognition and text-to-speech products, and also integrates technology from leading speech recognition providers. He is also our point person on industry standards, not only with the W3C but also with the IETF. (You can read the posts I’ve written about Dan’s standards work on our Speaking of Standards blog.)

As part of the VoiceXML Forum Board of Directors, Dan is looking forward to joining with the other board members in increasing awareness within the broader industry around what VoiceXML can do and also in expanding the certification efforts offered by the VoiceXML Forum.

We’re thrilled that Dan will be on the board and look forward to being part of the next stage of evolution of the VoiceXML Forum!

Are you looking for a free IVR platform to use on MacOS X or Linux? Would you like to build applications with open standards like VoiceXML, CCXML and SIP? Are you tired of all the solutions out there that require you to do all your development on Microsoft Windows?

Here at Voxeo we believe in Unlocked Communications™ and believe that you should NOT be locked in at any level – including the operating system!

That’s why we make our Prophecy platform available for all three major operating systems at:

www.voxeo.com/prophecy

And of course, Prophecy is free to download and use.[1]

Sure, the majority of our downloads reflect the market and are for Windows, but we use all three operating systems internally and have large customer deployments on all of them.

The best part of all of this?

Prophecy is the same on all three platforms.

And your apps are naturally movable between versions… so if your servers are all Windows or Linux but your developers all want to use shiny Mac laptops, they can install Prophecy locally on MacOS X and build all the apps they want. When it comes time to deploy, those apps can be moved over to your Windows or Linux server. Same platform, same tools, same code.

Unlocked from the operating system!

If you are interested in learning more, just head on over to our Prophecy download page and try out a version for yourself! Heck, it’s free, so feel free to download it onto several different systems and figure out which version you like best!

P.S. And just remember, the Linux version is the one that we showed CEO Jonathan Taylor demoing that runs over 5,000 simultaneous connections!

[1] The free version of Prophecy includes a license for 2 ports of simultaneous use. It’s the exact same code as the commercial version, though, so if you want more ports you simply have to contact us to buy a license for more.

This week at SpeechTEK in New York, we announced the general availability of Prophecy 10 with a 10X performance improvement, an improved user interface, new analytics, support for new audio codecs and so much more.  With “P10″, we had a unique opportunity to spend an entire release cycle focused on performance and reliability improvements given that our feature set is already so far ahead of our competition.  As we said in our news release:

Prophecy 10 is the result of more than 3,000 product performance, reliability, and usability improvements and investments made by Voxeo over the last eighteen months.  While IVR platforms built with VoiceXML and CCXML standards typically support no more than 500 concurrent calls per server, Voxeo’s intense focus on product quality improvements have given Prophecy the ability to support over 6,000 concurrent calls per server – more than 10-times the performance of other standards-based IVR platforms.  At the same time, Prophecy works perfectly well on low-end netbook, notebook, or embedded computers requiring anywhere from 1 to 100 concurrent calls.

We significantly slashed the server requirements… not just hypothetically… but in very real terms. One P10 server can literally replace 10 servers running our own previous releases or any of the current releases of competitors software.  This means less hardware, less power/AC costs, less administrative overhead… and on and on.  Much lower Total Cost of Ownership (TCO) and higher / faster Return On Investment (ROI).

Prophecy 10 is available now for free download for Windows, Linux and MacOS X at:

http://www.voxeo.com/prophecy10

As we hinted at in a previous post, we also brought a demo system to SpeechTEK that was running over 16,000 simultaneous IVR calls, with speech recognition included! With the P10 performance improvements, the system turned out to be rather easy to transport

I’ll be writing more about that system shortly and I’ve also recorded a video interview with Voxeo CEO Jonathan Taylor all about what we did. Stay tuned… it’s very cool!

Other features of Prophecy 10 include:

• Comprehensive analytics: Prophecy 10 bundles the proven capabilities of VoiceObjects Analyzer, including built-in VoiceObjects Analyzer reports for any existing VoiceXML application. Enterprises can access 30 additional reports by adding analytics notations to existing VoiceXML applications. VoiceObjects Analyzer also integrates with business intelligence (BI) solutions from SAP BusinessObjects, IBM Cognos and Microstrategy.
• New user interface – Prophecy 10 sports a new cleaner user interface for Prophecy Commander. We spent time working on usability to give you a cleaner, faster interface that lets you get to the information you need to administer the system as fast as possible.
• Intelligent transcription support: Unlike standard VoiceXML support for limited Automated Speech Recognition (ASR), Prophecy 10 uniquely includes support for intelligent speech transcription products and services from Nuance, PhoneTag, SpinVox, Yap, and Vlingo. This unique capability allows nearly any spoken dialog to be converted to text input, greatly widening the potential uses of speech-driven self-service.
• Proven call control: Prophecy supports advanced call control via both the W3C CCXML and Java SIP Servlet standards. Prophecy 10′s CCXML engine is the gold standard for XML-based call control. Prophecy’s built-in SIP Servlet engine is fully Java JSR-289 compliant and can support over 20,000 concurrent SIP sessions per server.
• Advanced enterprise and service provider multi-tenancy and virtualization: Platform resources can be logically separated, aggregated, or virtualized to partition customers or applications on a single server. Multiple servers and even multiple geographic installations can be combined to deliver extremely large multi-tenant systems. Prophecy’s multi-tenant capabilities are used to power the worlds largest and most widely-used XML telephony hosting services.
• Diverse speech recognition: Prophecy 10 includes free US English ASR and Text-To-Speech (TTS) engines and supports more than 30 additional language engines from Cepstral, IBM, Loquendo, LumenVox, Microsoft, Nuance and Telisma.
• RFC 5552: Support for SIP Interface to VoiceXML Media Services

Plus support for new audio codecs, new administrative features and so much more. We’re extremely pleased with how the release came out and we encourage you to download it today!

Today we are pleased to announce that Voxeo has acquired Motorola’s VoiceXML browser business, including full and exclusive rights to Motorola’s VoxGateway VoiceXML browser. Additionally, all existing OEM customers of Motorola’s VoiceXML browser are now Voxeo customers. As we’ve been talking about this announcement, we came up with the following answers to questions we’re being asked. If you have additional questions, please feel free to email me.

---

What is a “VoiceXML browser”?

A VoiceXML browser is the software you connect to when you call into a voice application. It is the interface between you as the caller and the automated application you are calling. If you think of a web browser on your computer, the interaction looks like this:

The web browser requests files from a web server and then presents them to you visually in the web browser window. A VoiceXML browser looks quite similar:

The main difference is obviously that the information retrieved is presented back to you as audio over your phone instead of visually on your computer screen. Here is a picture that shows the interaction in our hosted platform for an inbound call:

The voice browser is the software interacting with the caller.

What is special about Motorola’s VoiceXML browser?

Motorola had one of the very first commercial deployments of a VoiceXML browser back 10 years ago when VoiceXML first came out. The product, called “VoxGateway”, is one of the most widely used VoiceXML browsers with many companies licensing it. It is extremely modular and flexible for companies seeking to integrate it into existing products and is extremely standards-compliant. Some of the largest voicemail deployments in the world are built on VoxGateway.

What is the relationship between Motorola and Voxeo?

Voxeo was an early licensee of the VoxGateway software back in 2002. Once we had the source code, we began to build additional modules on top of the VoxGateway base. For instance, we added a module that used MRCP to interact with other speech engines and added a SIP module that let our Prophecy VoiceXML browser interact with VoIP systems. Today our Prophecy VoiceXML browser is an advanced… really supercharged VoiceXML browser built on the basic VoxGateway foundation.

In 2004, Voxeo entered into a partnership with Motorola where Voxeo took on responsibility for development of the basic VoxGateway product. Motorola still sold, licensed and supported the product, but Voxeo provided the development. Separately, Voxeo continued development of our advanced Prophecy VoiceXML browser. Along the way, we went to great lengths to ensure that Prophecy was the most VoiceXML-compliant browser on the market, passing all of the standard and optional VoiceXML compliance tests.

Today, we are announcing that Voxeo has acquired the complete VoxGateway business from Motorola, including all source code. We’ve licensed the software back to Motorola for their usage. The agreement also includes cross-licensing of related Voxeo and Motorola patents.

What does this mean for Motorola’s OEM customers?

Voxeo will be fully supporting all of the Motorola customers who have OEM’d the Motorola VXML browser. We will be continuing to maintain and develop the VoxGateway browser. The win for those OEM customers is that they can, if they choose, license the advanced modules we developed directly from Voxeo to get the added functionality.

What does this mean for Voxeo’s existing customers?

No immediate change, given that they are already using our advanced Prophecy VoiceXML browser. However, this strengthens our technology platform and gives us more options for the future.

Will Voxeo now be licensing the VoiceXML browser to other companies?

Definitely. For many years Voxeo has been licensing its technology to many of the other vendors in the industry. Our CCXML browser is the most widely licensed CCXML engine in the industry and is deployed in the networks of many of the vendors offering CCXML. Our speech engines (ASR and TTS) have been licensed out to other companies. In fact, our Evolution developer portal has been licensed and rebranded by companies seeking their own developer portal.

Now, with the full source code, we can license out to customers not only the basic VoxGateway browser but our full advanced Prophecy VoiceXML browser. As companies are increasingly looking to add VoiceXML functionality to existing products and services, we can now help them add VoiceXML without having to develop it all themselves.

Will Voxeo be enhancing the VoiceXML browser now that it owns it?

Certainly. The W3C is working on VoiceXML 3 now (and Voxeo’s Dan Burnett is actually co-chair of that effort), so you can expect to see us add that support as the standard matures. We have a lot of other ideas, too, so stay tuned…

We see it all the time – companies who deployed a standards-based voice application platform only to find out that they’re still locked in. How does that happen? After all, isn’t portability a key tenet of open standards?

Of course it is, but the fact is that your provider doesn’t want you going anywhere. To accomplish this, many vendors bundle proprietary tools that make it difficult to switch solutions should your requirements change down the road. Even standards-touting VoiceXML providers often use proprietary call control and VoiceXML tags to cleverly move lock-in from the platform level to the application level.

At Voxeo, we keep our customers by providing them with a better product and better support – never by locking them in. Voxeo offers the only VoiceXML browser to pass 100% of the VoiceXML Forum’s compliance test suite, including all optional tests, and the world’s most deployed CCXML browser. We support all of the secondary W3C voice standards and more SIP capabilities than any other vendor. Equally important, Voxeo strictly adheres to the principles behind open standards. The result is greater choice, unmatched flexibility, lower costs and true portability.

Since Voxeo’s customers own their intellectual property, they can switch providers, engage different development resources, or change their deployment model if requirements change.

With Voxeo you are free to:

• Move from hosting to a premise or hybrid deployment (or the other way around)
• Use your application development partner or VoiceXML development tool of choice
• Leverage the Voxeo VoiceObjects service creation environment to “design once and deploy anywhere.” VoiceObjects applications seamlessly work across all leading VoiceXML platforms and across multiple channels, including voice, IM, SMS and the mobile web. These phone channels can share the underlying dialog flow definition, business logic and back-end access, drastically simplifying development, maintenance, and reporting.
• Use your speech vendor of choice; Voxeo offers free built-in speech recognition and synthesis, as well as support for any MRCP-compliant engine
• Choose from a variety of TDM and VoIP carriers
• Run your application on any standards-compliant platform, including those of our competitors

If you are interested in learning more about Voxeo and our standards-based platform, contact us today.

As you are probably aware by now, at 6am US Eastern this morning / 12noon in Germany, we announced our acquisition of VoiceObjects, the world’s leading provider of self-service application development and analytics solutions. Given that over the course of the past several hours, we’ve been posting various articles and podcasts about the acquisition, I thought it was time for a summary of what we’ve talked about today. Here are some of the viewpoints we’ve shared:

• Our news release announcing the acquisition

• What the VoiceObjects acquisition by Voxeo means for voice application developers… on our Voxeo Developer’s Corner blog
• Voxeo’s VoiceObjects acquisition further promotes the open standard of VoiceXML on our Speaking of Standards blog
• Emerging Tech Talk #010 – Voxeo acquires VoiceObjects and brings developers new voice application tools on our Emerging Tech Talk video podcast

I think it goes without saying that we’re all extremely thrilled on both sides of the Atlantic Ocean by the combined potential of the two companies. In fact, we’re already seeing that potential be realized in some very huge opportunities that have come our way as a result of people within some large enterprises learning of the impending merger from one of the parties over the last few weeks. We’re looking forward to seeing what kind of even more amazing voice applications our customers will now create as we talk about and promote how well the tools of both companies work together.

From my own point-of-view, I look forward to joining with others within both companies in telling the story of how VoiceObjects application server and tools can be combined with Voxeo’s media server and application platform to solve real customer problems and more easily and simply build even more powerful voice applications. You can be sure that over the weeks and months ahead I and others will be writing more here (as well as on the VoiceObjects blog). It should be a very fun ride!

Meanwhile, if you haven’t watched our video introduction over on Emerging Tech Talk, let me make it even simpler for you by embedding the YouTube video here:

Keep watching for more… In the meantime, we’re celebrating a good bit both here in Orlando and over in Cologne, Germany! (As well as in our Beijing and London offices and wherever our other remote employees are…)

---

If you found this post interesting or useful, please consider:

• subscribing to the RSS feed for this blog;

• subscribing to the RSS feed for all of of our blogs;
• subscribing via email to the updates from all of our blogs; or
• following us on Twitter.

Thank you.

---

Technorati Tags: voxeo, voiceobjects, voicexml

UPDATE: In answer to several questions I’ve received, yes, I did record this presentation and will make that available soon. Once I do, I will also link it to the slide presentation so the slides can be viewed in sync with the audio.

---

As I mentioned previously (also here), I was at O’Reilly’s Open Source Convention (OSCON) last week up in Portland, Oregon, giving a talk on “Mashing Up Voice and the Web Through Open Source and XML“. It was fun to be back at OSCON for the first time since 2000 when I had been out there as the head of the Linux Professional Institute (LPI) and was talking then about Linux certification. It was great to see so many people who I knew from back in those days who are still very active in Linux and open source activities.

It was also quite frankly great to be out there not speaking about VoIP security but rather talking about voice mashups and XML! Don’t get me wrong, I still enjoy speaking about VoIP security and do it all the time (including next week at ClueCon) but as a speaker it’s sure nice to have some variety. It was also fun for me in this talk to dive down into the weeds and step through code line-by-line, something I haven’t done in quite some time in a presentation (but used to do all the time when teaching training classes).

Another fun aspect of the talk at OSCON was that Evan Prodromou, the founder and main developer of the open source identi.ca microblogging site was sitting there in the audience. Since most all of my demos involved adding voice to identi.ca, he was very interested in what I was showing and we had several great conversations later.

Anyway, my slides are now available through SlideShare, and I’d definitely welcome any comments or feedback on the slide set (either here or over on SlideShare) as I’ll be using some aspects of these slides in other talks:

I’ve received a good number of positive comments and evaluations about the talk and am glad that so many folks found it helpful. The one negative comment I’ve received was that a couple of people were really looking to see how they could add voice to their web interface and all my demos showed how to use a voice interface to interact with web sites. It’s a fair critique. I mention that you could do “click-to-call”, but I didn’t actually show this feature with a demo. I was originally planning to include a demo of putting a conference widget onto a web site, but in the end decided not to include it because I liked the other demos better and only had ~45 minutes for the whole presentation. Another time, though, I’ll probably include such a demo to provide both perspectives. (Meanwhile, I think that’s a great blog post to put up on our Voxeo Developers Corner blog in the next bit.)

Speaking of our Voxeo Developers Corner blog, I’ll also be taking each of my demos and stepping through them section by section in a series of posts over on that blog. Look for those to appear soon.

And, of course, if you’d like to try out any of the demos yourself using the code in my presentation, just head over to our Evolution developer portal and sign up for a free developer account and start creating applications.

P.S. Please do let me know if you do anything cool with linking voice into identi.ca (or Twitter) as it’s a subject I’m rather interested in. You can follow me there on identi.ca at identi.ca/danyork

Technorati Tags: SIP, slides, voice, voice applications, voice apps, voice mashups, voip, oscon, oscon2008, voxeo, mashups, voicexml, ccxml, identi.ca, microblogging, twitter, conferences, presentations, dan york, danyork

Over the next couple of months you’ll get a chance to hear from us and meet us at a couple of upcoming conferences.

First, from July 21-25 I’ll be up in Portland, Oregon, at O’Reilly’s OSCON 2008 giving a talk on “Mashing up Voice and the Web through Open Source and XML“. As the intro says:

With over 4.5 billion mobile and fixed phones out there as of November 2007, the phone represents the most ubiquitous user interface out there. As “mashups” on the web let us quickly and easily access information from multiple data sources, how do we extend those mashups to the world of the phone? How do we bring the old world of voice and telephony into the new world of the web, social networks and social media? And how do we do that using open source tools and open standards?

The session description gets into a bit more about specifically what I’ll be addressing. It should be a good time! As you can see from the schedule, OSCON is a strongly developer-focused show and so it should be a great time connecting with developers out there and talking about our platform. Speaking at this show is a bit of neat experience for me personally, too, as I spoke at one of the first OSCON’s back in 2000 when it was still called the “Open Source Conference” and was held in Monterey, CA. I know a good number of folks in the O’Reilly orbit so I’m looking forward to connecting with lots of people out there.

Next up, both RJ and I will be speaking at SpeechTEK, August 18-20 in New York City. In fact, we’ll have a team of Voxeons there as we’ll have an exhibit booth and all sorts of things going on at the show. (More on that in later posts.)

RJ will be speaking first on Monday, August 18th, along with VoiceXML Forum founder Ken Rehor on the subject of “The Impact of W3C standard languages“:

The publication of W3C standard languages, such as VoiceXML and CCXML, has dramatically changed the speech application design process. This session discusses some of the efforts to extend and validate the use of standard languages. Learn how the call control language can work with SIP and VoIP to implement an extensible SIP softswitch. Discover how the VoiceXML Forum’s certification program has impacted the cross-vendor interoperability of VoiceXML by VoiceXML platform vendors.

Given our huge focus on open standards and RJ’s direct involvement as chair of the W3C’s CCXML Working Group, you can expect this session to deliver a wealth of information! (Hmmm… how high can I set the expectation bar for RJ? Hmmm… “he’s going to show how open standards solve world hunger!”… no, that’s a bit too high! RJ is, though, an excellent presenter as you may get a glimpse of from his past slide decks.)

I’ll follow with a session on Tuesday, August 19th, called “Securing CCXML and VoiceXML Applications“:

How secure are your speech applications? As the usage of both VoiceXML and CCXML continues to explode, and VoIP usage continues to grow dramatically, especially within enterprise environments, it is increasingly important that you ensure that applications and services are not open to attack. Learn about the potential vulnerabilities in a system using VoiceXML or CCXML, what you can do to secure these systems, and how you can develop a strong architecture.

Given my VoIP security background, it’s somewhat predictable that I’d be talking on this subject, eh? Seriously, though, it’s an area that doesn’t seem to be getting a whole lot of thought and so I’ll be taking a look at what the real risks are and how you can look at addressing them.

If you are going to either OSCON or SpeechTEK, please do leave a comment here or drop us an email and let us know. We love to meet face-to-face with blog readers and/or customers/developers. As much as we are able to do so (not sure yet), we’ll make the presentations available through this site. Stay tuned for more info – and we look forward to potentially seeing some of you in either Portland or New York!

Technorati Tags: voxeo, conferences, rj auburn, dan york, oscon, oreilly, oscon2008, speechtek, speechtek2008, voicexml, security, voip security, ccxml, mashups, xml

One of the fun aspects of giving your product away for free is that you really have no clue what people will wind up doing with it. So we were delighted to learn that our premise “Prophecy” product as well as industry standards we support like VoiceXML and grXML were included in a new book, “Mobile Web Development“, new available from Packt Publishing. (UPDATE: I subsequently learned that the author did contact us some time back about permission to use some screenshots, etc.)

The book’s author has a blog and in his entry today about the book writes this (I obviously added the bold emphasis):

Mobile Web Development shows you how to build a mobile presence for your web applications and sites. It covers targeting different mobile web browsers, sending and receiving SMS and MMS messages, accepting mobile payments, and developing voice- and touchtone-response systems.

The book illustrates every technique with practical examples, showing how to use these development methods in the real world. Along the way we show how an example pizza delivery business can use these methods to open up to the mobile web. The book is fast paced, yet touches upon relevant concepts. I had a great ride while I was reviewing the book. I was actually surprised at how fast the book moved – covering important topics in just enough detail, and moving on to the next!

There is a whole lot of technologies I covered in the book. The examples are done in PHP with XHTML Mobile Profile. But I also covered: AJAX, Wireless CSS, WAP 2.0, Tera WURFL, MyMobileWeb, Mobile Web Toolkit, Image Server, GAIA Image Transcoder, HAWHAW , ready.mobi, Nokia’s Mobile Internet Toolkit , SMIL, Voxeo’s Prophecy server, VoiceXML (VXML), grXML, Frost library, WALL, WURFL, SMS, MMS, mobile payment gateways, mobile widgets, Mobile AJAX, Android, Google Gears, and Dojo Offline.

The blog entry goes on to talk more about what the book offers developers. It sounds like a great book and I’m looking forward to seeing a copy (since I actually have no idea what he said about us!). We do see the whole mobile space as a huge growth area and welcome developers to look at using either our on-premise Prophecy or our hosted Evolution platform to build their apps. (Check out the docs available at either of those links!)

Technorati Tags: books, mobile web development, nirav mehta, voxeo, prophecy, voicexml, vxml, grXML, mobile

We are pleased to announce the launch of a new weblog: “Voxeo Developer’s Corner“. In this weblog we will be giving you tips, tricks and tutorials to help you develop applications on our platforms. As you are probably aware, there is a huge amount of information available in the documentation and forums found on our Evolution developer site. What our goal is with this blog site is to point people to some of the nuggets of gold buried within the pages of Evolution. Members of our staff will highlight some of the great tips they have found and other information they think might help you build even better voice applications on our platform.

The blog is off to a great start already with Matt Henry’s initial post, “Certified Tech Tip: Using SISR-formatted grammars with Prophecy 8“, where he demonstrates what SISR-formatted grammars are all about. The post attracted some comments and Matt is already planning a follow-up post. Expect to see more along these lines from Matt and his team in the weeks ahead.

I also kicked off a series on voice mashups and Twitter today with a post “Voice Mashups with Twitter, part 1: Who will win the 2008 SuperBowl? (A mashup in CallXML.) where I show how you can create a voice application that sends its output out to the Twitter micro-blogging service. Over the next few weeks I’ll be adding more articles in that series.

If you are interested in developing voice applications, we do encourage you to check out the “Voxeo Developer’s Corner“. There’s also an RSS feed if you would like to subscribe and stay up-to-date with our posts.

Technorati Tags: applications, voxeo, developers, callxml, xml, voicexml, mashups, ccxml, apis