speech-recognition

问题 I wrote an app in C# for speech recognition using System.Speech which works fine on Windows 7. However I'm after creating the same app that will work on windows 2003 (x86). My programming environment: Windows 7 x64 Pro Visual Studio 2008 In order to develop this application in my programming environment I installed: 1.Microsoft Speech Platform - Server Runtime (Version 10.1) (x86) http://www.microsoft.com/downloads/details.aspx?FamilyID=674356C4-E742-4855-B3CC-FC4D5522C449&displaylang=en

问题 I wrote an app in C# for speech recognition using System.Speech which works fine on Windows 7. However I'm after creating the same app that will work on windows 2003 (x86). My programming environment: Windows 7 x64 Pro Visual Studio 2008 In order to develop this application in my programming environment I installed: 1.Microsoft Speech Platform - Server Runtime (Version 10.1) (x86) http://www.microsoft.com/downloads/details.aspx?FamilyID=674356C4-E742-4855-B3CC-FC4D5522C449&displaylang=en

问题 i am working on an android application using RecognizerIntent.ACTION_RECOGNIZE_SPEECH,,, my problem is that i don't know how to create the buffer which will capture the voice that the user inputs. i read alot on stack overflow, but i just don't understand how i will include the buffer and the recognition service call back into my code. AND HOW WILL I DO PLAY BACK FOR THE CONTENTS WHICH WERE SAVED INTO THE BUFFER. this is my code: public class Voice extends Activity implements OnClickListener

问题 i am working on an android application using RecognizerIntent.ACTION_RECOGNIZE_SPEECH,,, my problem is that i don't know how to create the buffer which will capture the voice that the user inputs. i read alot on stack overflow, but i just don't understand how i will include the buffer and the recognition service call back into my code. AND HOW WILL I DO PLAY BACK FOR THE CONTENTS WHICH WERE SAVED INTO THE BUFFER. this is my code: public class Voice extends Activity implements OnClickListener

问题 I have constructed a GMM-UBM model for the speaker recognition purpose. The output of models adapted for each speaker some scores calculated by log likelihood ratio. Now I want to convert these likelihood scores to equivalent number between 0 and 100. Can anybody guide me please? 回答1: There is no straightforward formula. You can do simple things like prob = exp(logratio_score) but those might not reflect the true distribution of your data. The computed probability percentage of your samples

问题 I try to run a demo for speaker verification using MSR Identity toolkit. However it left error after training UBM step. The error is as follow. It looks like fopen return -1 and cause error to fread . I can't understand why it can't read the filenames . I can't attach the code since it involves many functions. I just hope someone that familiar with this toolkit can help me. Error using fread Invalid file identifier. Use fopen to generate a valid file identifier. Error in htkread (line 7)

问题 I'm trying to inject the transcribed text in the search query in order to filter my adapter list, but, after trying every combination I could imagine, I come here to ask for help. For instance, if handle the intent after a new intent and set text to a TextView everything works fine and the voice recorder transcribes what is said correctly, but I'm not being able to get the same String and set the SearchView query successfully. The keyboard microphone, otherwise, captures the text and pass it

问题 In my Speech Engine I activate / desactivate multiple grammars. In a special step I'd like to run a Grammar with ONLY to capture Audio of next given sentence according to engine's properties. But to start/stop matching something, I assume engine needs "words". So I don't know how to do it ? (Underlaying explanation: my application convert all garbage audio to text using google speech API because dictation is too bad and no available on Kinect) 回答1: Well, actually, no, the SR engine only needs

问题 In my Speech Engine I activate / desactivate multiple grammars. In a special step I'd like to run a Grammar with ONLY to capture Audio of next given sentence according to engine's properties. But to start/stop matching something, I assume engine needs "words". So I don't know how to do it ? (Underlaying explanation: my application convert all garbage audio to text using google speech API because dictation is too bad and no available on Kinect) 回答1: Well, actually, no, the SR engine only needs

问题 I am working with Speech Recognition in Swift and have a problem with canceling SFSpeechRecognitionTask . I have implemented the following logic: I have a mic button Speech is being recognized while button is pressed When the mic button is released I need to finish the speech recognition, to get the results in completion block ( recognitionTask is used for it). So I use stopRecording() function for this purpose. This function works properly. I also need to cancel recognitionTask when I do