Is it possible to create a grammar file for speech recognition and use SpeechRecognizer offline, like Cortana?
I only found Speech.Recognizer examples... it seems like Recognizer can use SpeechRecognitionEngine.loadGrammar(). I wonder if something similar is possible on a Windows Phone?
Is it possible to use SpeechRecognizer offline with a grammar File? Or is it possible to use Cortana's ability to recognize Speech To Text offline?
Related
Background info
I am trying to convert a microphone recording to a string.
So far I have been following this sample from MSDN website to create a program that allows me to record with the microphone and play it back. Below is the link.
http://msdn.microsoft.com/en-us/library/windows/apps/gg442302%28v=vs.105%29.aspx
Sadly, I have not found any way how I can convert recording to string.
Question How do I convert the microphone recording to string?
I have tried to use Speech Recognition Engine but unfortunately the namespace System.Speech cannot be used in Windows Phone 8 sdk.
You want Windows.Phone.Speech for speech recognition, not System.Speech. An overview is here.
I want to make a program using Windows speech recognition and the respective .NET API and I was wondering, will the program automatically use the grammar and all the data that has been collected from the training of the recognition engine? Or will I have to force it? And if it has to be forced, how will that be ahcieved? The language to be used is C#.
The training is used so Windows Speech Recognition better understands what you say and gives more accurate results. This training data is stored in profiles. When you use Windows Speech Recognition in C#, you don't have to select a profile, Windows does this automatically.
I want to know how can I add commands to windows 7 shared speech recognition using SAPI 5.4 in C#. There is already a Microsoft application named WSRMacros, but I should do it programmatically by myself.
Any help and explanation would be extremely appreciated.
Just create an application that uses System.Speech.Recognizer.SpeechRecognizer. This loads the shared recognizer, which starts Windows Speech Recognition.
Create your grammars, and off you go!
I have a Windows Phone 8 app that uses the SpeechRecognizerUI package to do speech recognition. It works fine, but with every recognition session, the speech recognition engine plays a couple "bloop" style sound effects. I'd like to get rid of them, but I don't see anything in the Settings property that lets me turn those off. Does anyone know how to do this?
Note I already have the Settings properties ReadOutEnabled and ShowConfirmation set to FALSE.
The sound effects are part of the SpeechRecognizerUI, if you don't want to have them you can use SpeechRecognizer instead and build your own UI.
I tried implementing some simple speech recognition WinForms program in C# like the one described here in Michael Levy answer:
good Speech recognition API
The problem i have is that any time i run the program Windows Speech Recognition opens and is also doing stuff based on what i am saying. Also when the program starts i have to say "start listening" for speech recognition to work.
My question is: How can i use speech recognition without having Windows Speech Recognition also act on what i am saying? I don't need Windows Speech Recognition UI to open at all and i need to be able to use recognition without having to say "start listening" before.
Thanks for your answers
Are you sure you are using an inproc recognizer for your application only. You do this by instantiating a SpeechRecognitionEngine() in your application. See SpeechRecognitionEngine Class. I suspect you are instantiating a shared recognizer - SpeechRecognizer Class