Kevin Junghans

I had a couple of enhancement requests in the VoiceModel Project around making user recordings.  The first request was to allow the directory that holds the recordings to be specified as a file structure or file share. And the second request was to allow multiple recordings in a single session.  I finally got some time to look at implementing these enhancements and was pleasantly surprised that the features were already in place because of the flexibility in the VoiceModel architecture and state machine. Well, for the most part. I will explain one little change I had to make to VoiceModel to get this to work. To test this out I updated the RecordingExample project in the VMWithExamples solution. You can get the source code here . Basically my solution was to add an On-Exit-Action to the recording state that copies the recorded file to another location.  This satisfies the request to be able to put recording in another location, including a file share.   Here is ...

A new version of VoiceModel has been released that allows your voice applications to run on the Tropo Cloud IVR . Now you can develop your voice applications once and run them on any VoiceXML compatible IVR or on the Tropo Platform.  VoiceModel is an open source project and you can find examples on how to develop with VoiceModel in the source code .

In my last post I discussed how to save a recording from Tropo using ASP.NET MVC and VoiceModel . I also discussed one of the issues I had to deal with was a difference in how VoiceXML and Tropo handle recordings.  A nice feature in VoiceXML is that you can listen to the recording on the IVR before you save it. With Tropo's Web API there is no mechanism to do this.  The goal of VoiceModel is to develop your voice application once and you can run it on either a VoiceXML or Tropo platform,  therefore I had to make both behave in the same manner.  The solution was to save the recording immediately on completion and to let the user listen to the recording on the application server that VoiceModel is deployed on. In order to grab the correct recording from the application server I needed a method to map a recording to the current session. Fortunately both VoiceXML and Tropo send a unique session ID with each request so I used the session ID as the file name for the audi...

We are making progress on the Tropo integration in VoiceModel .   This will allow applications developed in VoiceModel to run as VoiceXML applications or Tropo applications without any changes.  All you have to do is change your application URL in the platform configuration.  So if you deploy your application at http://myServer/myApp  and  have a VoiceModel controller called myController then the starting URL set in Tropo would be http://myServer/myApp/my/StartTropo .  That is all there is to getting your Tropo application running. We are going through the sample applications and using them as a test bed.  Just finished the WeatherVoice example which asks the caller to enter their zip code, retrieves the weather conditions for the area from a service, and voices the data back.  Give it a try by getting the latest version in source control . You can run WeatherVoice on either Voxeo's Prophecy IVR  (in the cloud or on premise), on any VoiceX...

It looks like Google has killed its unofficial weather service API.  I caught wind of it when I saw a lot of inquiries about it on StackOverflow .  Now when you make a request to the API it returns "Unsupported API".  One of VoiceModels examples used the Google weather API to demonstrate how to get information from a service and voice it back.  When I found this out I searched for another free weather service and found MSN. The example WeatherVoice now uses the MSN service.  The example was written so that it is easy to insert new services or even use a mock-up service when testing.  If you need a replacement for using the Google weather API you may want to check out the example to see how you can use MSN for the same information.

I recently used VoiceModel to create an add-on module for Mayhem , an open source project developed by Microsoft that allows non-programmers to use their computers to automate anything.  You can find out more about Mayhem here and about the contest that is running until the end of the month for contributing add-on modules.  I have submitted the add-on module called RemoteCommand which makes it easy to integrate Mayhem with an IVR application.  Watch this video to see how Mayhem, RemoteCommand, and VoiceModel can be used together to automate actions through commands given over the telephone. The voice application used in this demonstration can be found in the VoiceModel examples  found in the source code.  This example demonstrates how to dynamically build speech grammars with VoiceModel. I plan on using this for home automation by creating another Mayhem add-on module for Z-Wave devices.  This will allow for controlling Z-Wave devices i...

I will be giving a presentation on how to create an open source reusable framework using ASP.NET MVC to the local .Net Developer Group in Sarasota, Florida on June 20, 2012.  In this presentation I will be covering the architecture behind VoiceModel , an open source project available on CodePlex that makes it easier to develop VoiceXML applications using ASP.NET MVC, Razor and C#.  If you are in the area be sure to drop by.  The exact time and directions are available here .  I think you will find the presentation very informative. And as always after the meeting there will be continued discussions and networking at the Cock & Bull . They have the greatest beer selection in Sarasota, if not all of Florida.  So come and join in on a fun and informative evening.

VoiceModel does not contain a Menu object that will present users with options and branch in the call flow based on the users input.  So how do you add a menu to your voice application when using VoiceModel? Easy. Just add an Ask object and put conditions on the transitions for the state that renders the Ask object to the IVR browser.  Here is some sample code that you can find in the examples for VoiceModel . public class HomeController : VoiceController { public override CallFlow BuildCallFlow() { CallFlow flow = new CallFlow(); flow.AddState(ViewStateBuilder.Build("mainMenu", new Ask("mainMenu", "Press one for option one. Press two for option two.", new Grammar("digits?maxlength=1"))) .AddTransition("continue", "optionOne", new Condition("result == '1'")) .AddTransition("continue",...

I have been doing some work with workflow engines lately and it got me thinking; what is the difference between a workflow and a state machine? Further reflection had me wondering whether a voice application call flow is best represented by a state machine or a workflow. Discussions I had while working with the VoiceXML Forums Tools Committee raised a third option of using rules engines for more complex applications.  So what is the best option for implementing a voice application? To determine this lets examine what each method of modeling application logic actually is.  But first lets make sure we are on the same page as to what a voice application call flow is. Voice/IVR application call flows are usually designed on paper prior to implementation. This design shows the possible paths a user can take through the voice application depending upon their responses and data that is returned by external and internal systems. A common method of depicting call flow is to use...

UPDATE: How Reusable Dialog Components in VoiceModel are created and used has been greatly simplified. Go to this post for more details . This post is a continuation of a previous post on using Reusable Dialog Components (RDC).  In this post we will look at creating Reusable Dialog Components for a VoiceXML application using VoiceModel .  VoiceModel is an open source project that makes development of VoiceXML applications easier using ASP.NET MVC, C#, and Razor.  Reusable Dialog Components are reusable libraries that make development of VoiceXML application easier as discussed here .  In this example we will create an RDC for getting a date from a user using a telephone keypad. To get started create a class library project in Visual Studio.  VoiceModel is developed in Visual Studio 2010 using .NET Framework version 4.0.  Next include in the references System.Web , System.Web.Mvc , VoiceModel , and MvcContrib .   MvcContrib is available via NuGet ...

In a previous post I discussed what Reusable Dialog Components (RDC) are and the basis for adding support for them in the VoiceModel Project .  In this post I will discuss how to use RDC's with  the VoiceModel framework, which allows developers to create advanced voice applications using ASP.NET MVC, C# and Razor.  This will walk you through the example for RDC's in the VoiceModel Project that you can download from here .  Download the complete project for version 0.5 on this page and you will find this example in the project directory ReusableComponentEx . To get started create a new empty MVC 3 project in Visual Studio.  Next, include a reference to the VoiceModel assembly and the RDC assembly.  The RDC assembly is called GetDateDtmf .  This RDC provides a dialog for collecting a date from the user using the telephone keypad.  VoiceModel uses a technology called Portable Areas to be able to reuse Controllers, Views, and other resources that ...

Reusable Dialog Components (RDC) are an important part of making development of VoiceXML more productive.  There are portions of any VoiceXML application that are often repeated and therefore like any application development environment we need a modular way to reuse these components.  And often they are repeated across applications. Take for example a dialog to get a date from the user. There are certain common steps to get the date, validate it, confirm that is the date the user intended, and retry if it is invalid or not the intended date.  Rather than repeat the code in your application to get a data over an over again it is much more productive to call a component library that contains this routine.  RDC's can be as simple as getting a date from a user or more complex and build on each other.  For example, an RDC for getting credit card information could be built on smaller RDC's for getting the date (date of expiration), credit card number, and security co...

Recording user input in a VoiceXML application using ASP.NET MVC is very easy if you use VoiceModel .  VoiceModel is an open source project that abstracts the VoiceXML to an easy to use object hierarchy that is meant to work with ASP.NET MVC 3 and Razor.  In this post I will walk you through a simple demonstration on using VoiceModel to record user input.  The source code for this demo is available on CodePlex .  In this demonstration the application will prompt the user to make a recording after a beep and to press the pound (#) key when they are done. When the user is finished recording the recorded message will be played back to them and they will be asked if they want to save it. If they respond yes, by either saying "yes" or pressing the 1 key on the phone, the audio recording will be saved in a directory on the application server.  If they respond no, by saying "no" or pressing the 2 key on the phone, they will be prompted to re-record the mess...

I was working on adding the ability to record user input in VoiceModel  when I ran into an issue with the syntax of the VoiceXML that was returned to the IVR.  I was testing the changes using Voxeo's Prophecy IVR  and the logs told me what line the issue was on but I could not understand why my application was generating the VoiceXML in question and I could not see VoiceXML document to verify what was going on.  My usual step for troubleshooting issues like this is to take the URL from the Prophecy log associated with the issue and put it in a web browser to see what the IVR is working with.  But in this case the IVR was doing a POST and a web browser can only perform GET operations. After you create a user recording in VoiceXML using the record element you need to send the audio file from the IVR to you application using the submit element in order to save it, as shown below.  The recording does not do you much good if you leave it on the IVR. <sub...

In this post I will show you how to use the Google Weather API to retrieve current weather conditions that can be voiced back to callers.  In a previous post I showed how to use VoiceModel to create a simple VoiceXML application that asks the caller for their zip code, looks up the the weather conditions for that zip, and then voices the information back to the caller; but I used a mock-up for the service that retrieves the weather information for easier testing.  This time I will demonstrate how to swap out the mock-up for a functioning service to get the current weather.  Since the underlying architecture for VoiceModel is ASP.NET MVC this method would work just as well in a web application. The way I implemented the Weather Application the service to retrieve the weather information implements the interface IWeatherService . This allows me to easily insert a mock-up for testing or any version of a service that I wish to.  I chose the Google Weather API because ...

This post discusses some of the design decisions used for developing VoiceModel , an open source project for creating a framework that simplifies developing VoiceXML applications using Microsoft ASP.NET MVC and Visual Studio. Developing voice/speech applications using VoiceXML has made development of IVR applications more like developing web applications.  The IVR acts like a web browser.  In fact the component of the IVR that consumes the VoiceXML documents is often referred to as the VoiceXML browser.  So that would make the IVR the client-side.  Any application logic we put into the VoiceXML document will be processed client-side.  Even though some constructs were added in later versions of VoiceXML to make it possible to access data sources directly from VoiceXML you usually needed server-side programming to access back-end systems to retrieve or update data and to dynamically generate VoiceXML documents that contain data to be voiced back to...

I found a couple of issues when deploying a VoiceXML application using MVC and VoiceModel  to a hosted environment and I wanted to share the solutions here. First you may get an error message when you try to run the application that it cannot find certain assemblies that are required for MVC.  In an earlier post titled " Using ASP.Net MVC to Create a Simple VoiceXML Application " I discuss how you have to install MVC 3 because it does not come with VS 2010 right out of the box.  Well it is currently not part of the .NET Framework either.  But there is an easy solution for including the correct assemblies as described in this blog post .  Just right click on your project and select Add Deployable Assemblies .  I dialog box will appear and select the option for ASP.NET MVC .  Now when you publish your project the appropriate assemblies will be deployed with it. The other issue is that VoiceModel deploys shared views in the ~/tmp/Views/ directory. ...

In a previous post we created a simple "Hello World" application to demonstrate how to use the VoiceModel library.  This did not demonstrate how to control the flow of a voice application so in this post I will explore how we can add a controller to our MVC VoiceXML application.  Lets take a simple example of a voice application that allows the caller to get the weather conditions for an area based on zip code.  Here is the call flow for our application. If we were developing this application in straight VoiceXML the first block in our call flow could be handled something like this. <?xml version="1.0" encoding="UTF-8"?> <vxml xsi:schemaLocation="http://www.w3.org/2001/vxml http://www.w3.org/TR/voicexml20/vxml.xsd" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns="http://www.w3.org/2001/vxml" version="2.1"> <form id="greeting"> <block> <prompt...

This is a first in a series of blogs where I will explore using new Microsoft .Net technologies to develop voice/speech applications that will run on any IVR platform that supports VoiceXML.  In this initial blog I will take a brief look at the history of voice application development and why I think this is a viable and productive method for developing voice applications. I will also introduce you to the specific technologies I plan on using. I have been developing voice/speech applications for over 20 years and a lot has changed over these years.  The biggest change was from developing in proprietary development environments supplied by each IVR vendor to IVR platforms that supported open standards like VoiceXML and CCXML  that were controlled by the World Wide Web Consortium (W3C).  These open standards radically changed how voice application were developed, moving to more web-centric tools and promises of making every web developer a voice application develop...