i want to get started with speech recognition and speech synthesis
in Prototype based on speech recognition
somebody told me to use microsoft speech server (sdk and so on)
when i have this, how do i programm an application, and with which programming language (development enviroment?)?
has someone experience with asterisk or SVOX?
i need to do:
speech recognition
speech synthesis
i doesn't have to be a very good speech recognition - i think 30 -50 words should be enough for the beginning.
i'm working with windows.
thanks in advance
If you choose to use the Microsoft Speech Engine, there are .NET framework APIs. As I mentioned in the other post, there are two namespaces (System.Speech for desktop use and Microsoft.Speech for server use). You can program in any .NET language and you can use Visual Studio.
There is a very good article that was published a few years ago at http://msdn.microsoft.com/en-us/magazine/cc163663.aspx. It is probably the best introductory article I’ve found so far. However, it was based on a prerelease version of the WinFX API and the System.Speech classes were changed when Vista was released. The samples in the article do not compile because of these breaking API changes and I have not found any updates or errata to explain this. Search the Internet for the method name “AppendResultKeyValue”, you’ll find a few forum posts like http://www.ms-news.net/f3012/system-speech-breaking-changes-3025734.html where people ran into this same problem.
It is still a good introductory article and well worth reading. with a little bit of hacking, you can get the sample app working.
When I studied computational linguistics the tool of choice was Praat, a horribly confused prototyping tool that lets you do just about anything speech-related.
I don't think it has any external API, but its internal scripting language is sufficient for rudimentary applications, and it has lots of built-in functions. For "getting started" on theory and algorithms it is not too bad.
