tr00p3r
Newbie
Posts: 3
Joined: Tue May 19, 2015 8:25 am

Benefits over System.Speech?

Tue May 19, 2015 2:01 pm

Hi,

I've got an app written using System.Speech and at times I'll experience a crash due to stop and starting the speech recognizer and the recognition isn't that great.

Is CMUSphinx a better alternative to System.Speech or is converting over to Syn.Speech not going reap any rewards?

User avatar
Leslie
Lead Software Architect
Posts: 377
Joined: Fri Sep 14, 2012 12:20 pm
Contact: Website

Re: Benefits over System.Speech?

Tue May 19, 2015 11:42 pm

Syn.Speech and CMUSphinx share the same architecture. Their recognition should, in theory, be similar.

If your app has crashed for some reason then do share the StackTrace so we can fix the issue/bug immediately.

Coming to benefits, it's all about the Language and Platform of choice. If you are working with Java I myself recommend you sticking with CMUSphinx.

tr00p3r
Newbie
Posts: 3
Joined: Tue May 19, 2015 8:25 am

Re: Benefits over System.Speech?

Wed May 20, 2015 6:11 am

Sorry, I meant Microsoft's System.Speech vs CMUSphix.

User avatar
Leslie
Lead Software Architect
Posts: 377
Joined: Fri Sep 14, 2012 12:20 pm
Contact: Website

Re: Benefits over System.Speech?

Wed May 20, 2015 10:14 pm

Well, my bad I should've paid attention.

I have to be honest, when I am dealing with restricted Grammar I have always preferred System.Speech over CMUSphinx. It simply works. There's one major caveat to using CMUSphinx or Syn.Speech (at the moment) is that you cannot train these speech recognition engines. On the other hand the Static Speech Recognizer in System.Speech can use the user's speech profile to yield better results.

If you are targeting Windows/Mac then I reckon it's better sticking with System.Speech.

Return to “General Discussion”

Who is online

Users browsing this forum: No registered users and 1 guest