Q: Does Speechdraw require a proprietary platform to work?
No. SpeechDraw generates static VXML pages that can run on any platform that is VXML 2.0 compliant.
Q: Is it possible to create simple IVR applications using SpeechDraw?
Yes. Even if Speechdraw is conceived to design complex mixed-initiative dialogues, it can also be used to design traditional touchtone IVRs, and faster than with traditional tools.
A further advantage is that the same application can be easily be transformed in a full speech application.
Q: Being SpeechDraw a graphical tool, how can programming be used to solve special tasks?
Speechdraw enables to insert javascript coding in any point of the dialogue. Using the built-in jscript editor, it is possible to manage all the environmental variables in a very easy way.
Besides that, it is possible to access to external pages to define the processing of any variable used in the dialogue.
Thanks to this mechanism, Speechdraw enables you to do any kind of post processing, for example, semantic post processing, complex natural language understanding, etc.
Q: Is it necessary to be a Vxml programmer to develop speech applications using SpeechDraw?
Absolutely not! One of the key features of SpeechDraw is that it enables to develop complex speech applications with zero programming effort.
First class applications, with mixed-initiative dialogues, multislot grammars, adaptive prompts, etc. can be done without any knowledge of Vxml programming and of grammar syntax.
Q: Is it difficult to use SpeechDraw?
We claim that Speechdraw is the easiest tool to develop applications. With few hours of training, you can start to develop a real speech application. An automatic assistant will check throughout the development of the application and point out any errors. The assistant will then suggest the solution and show where the error occurred.
The assistant will allow you to save many hours of debugging!
Q: Is it possible to use reusable objects in SpeechDraw?
With Speechdraw you can transform any application into a reusable object. A customizable interface can be defined, setting all the parameters that will control the behaviour of the object. Reusable objects can be nested with no limits.
You can create a library of reusable objects to dramatically speed up the development.
Q: Can Speechdraw manage large projects ?
In Speechdraw, you can define subdialogue objects (subdiagrams can be nested) to split complex applications. What is even more important is that, even if you started your application as a big unique block, you can at any time divide your application in subdialogues. This makes the flow much more readable and maintainable.
Q: Can Speechdraw be used in a network environment and with large teams ?
Speechdraw is conceived for large teams. The administrator can define teams and associate different teams to different projects. It is even possible to define a profile for each designer of the team and to specify which activity a specific designer can do.
Q: Does SpeechDraw generate any documentation ?
SpeechDraw automatically generates a full documentation of the application.
The documentation reports the details of the application designed: the generated flows, all the prompts, the structure of the grammars, etc.
Reading the documentation, it is easy to understand the logic of the application and all the details of the error recovery defined.
Q: Is it possible to design grammars using SpeechDraw ?
Speechdraw has a powerful built-in grammar builder. The grammar builder has a graphical environment that enables to develop grammars in a fraction of the time required using the traditional approach, even if the designer is a novice in grammar design.
Q: Is the grammar builder limited to the design of simple grammars or can it be used for more complex ones too ?
With the grammar builder you can design complex multislot grammars.
You can design complex linguistic graphs with distributed semantics.
For each node of the grammar, it is possible to define semantic processing, as well as global processing for the entire grammar.
Spelling, concatenations of strings, calculations etc. can be very quickly defined.
With the grammar builder you can even define logical implications between distant concepts.
External grammars can be inserted at any point with a simple drag&drop from the libraries.
Q: How can SpeechDraw be used to access a database?
You can access a database using the mechanism of the host access service API definition: from inside Speechdraw you define the parameters that must be sent and returned during the exchange with the backend.
Once defined the API, SpeechDraw automatically generates the Vxml frame to connect to the host.
Q: Since SpeechDraw is not bound to any specific application server, how is it possible to run applications without a preexisting integration with the backend?
With Speechdraw, you can "emulate" the backend. Using the emulation you can test the application logics even if no integration is ready.
This is fundamental for fast prototyping.
Once the integration is ready, provided it has met the specifications set form the host access API definition, it is sufficient to compile the application to make it run in "real" mode instead of "emulation" mode.
The API specifications - for the services to be developed - are reported in the documentation generated by SpeechDraw.
Q: Does SpeechDraw have any mechanism to analyze the log of the calls?
SpeechDraw has a log analyzer that analyzes the calls related to the application, and generates a statistic of the results.
The log analyzer correlates the analyses of the calls to the application flow, showing the performance of each element of the user interface.
For example, for each block of the flow, the log analyzer reports if the block did recognize what the user said; if it recognized with good performance, with or without error recovery etc.; how deep the escalation of the error recovery used was, to recover any errors; how the various turns of the dialogue contributed to the overall duration of the call; etc.
The log analyzer is fundamental to understand how to improve the application , during the tuning phase.
