This commit is contained in:
2017-09-26 20:21:42 +02:00
parent e144bde202
commit 319ca0cbbb
194 changed files with 10149 additions and 0 deletions
+20
View File
@@ -0,0 +1,20 @@
Microsoft Visual Studio Solution File, Format Version 11.00
# Visual Studio 2010
Project("{FAE04EC0-301F-11D3-BF4B-00C04F79EFBC}") = "Demo3", "Demo3\Demo3.csproj", "{33D30437-2141-4BEE-B4DE-FBF5B01F7A8C}"
EndProject
Global
GlobalSection(SolutionConfigurationPlatforms) = preSolution
Debug|x86 = Debug|x86
Release|x86 = Release|x86
EndGlobalSection
GlobalSection(ProjectConfigurationPlatforms) = postSolution
{33D30437-2141-4BEE-B4DE-FBF5B01F7A8C}.Debug|x86.ActiveCfg = Debug|x86
{33D30437-2141-4BEE-B4DE-FBF5B01F7A8C}.Debug|x86.Build.0 = Debug|x86
{33D30437-2141-4BEE-B4DE-FBF5B01F7A8C}.Release|x86.ActiveCfg = Release|x86
{33D30437-2141-4BEE-B4DE-FBF5B01F7A8C}.Release|x86.Build.0 = Release|x86
EndGlobalSection
GlobalSection(SolutionProperties) = preSolution
HideSolutionNode = FALSE
EndGlobalSection
EndGlobal
+63
View File
@@ -0,0 +1,63 @@
<?xml version="1.0" encoding="utf-8"?>
<Project ToolsVersion="4.0" DefaultTargets="Build" xmlns="http://schemas.microsoft.com/developer/msbuild/2003">
<PropertyGroup>
<Configuration Condition=" '$(Configuration)' == '' ">Debug</Configuration>
<Platform Condition=" '$(Platform)' == '' ">x86</Platform>
<ProductVersion>8.0.30703</ProductVersion>
<SchemaVersion>2.0</SchemaVersion>
<ProjectGuid>{33D30437-2141-4BEE-B4DE-FBF5B01F7A8C}</ProjectGuid>
<OutputType>Exe</OutputType>
<AppDesignerFolder>Properties</AppDesignerFolder>
<RootNamespace>Demo3</RootNamespace>
<AssemblyName>Demo3</AssemblyName>
<TargetFrameworkVersion>v4.0</TargetFrameworkVersion>
<TargetFrameworkProfile>Client</TargetFrameworkProfile>
<FileAlignment>512</FileAlignment>
</PropertyGroup>
<PropertyGroup Condition=" '$(Configuration)|$(Platform)' == 'Debug|x86' ">
<PlatformTarget>x86</PlatformTarget>
<DebugSymbols>true</DebugSymbols>
<DebugType>full</DebugType>
<Optimize>false</Optimize>
<OutputPath>bin\Debug\</OutputPath>
<DefineConstants>DEBUG;TRACE</DefineConstants>
<ErrorReport>prompt</ErrorReport>
<WarningLevel>4</WarningLevel>
</PropertyGroup>
<PropertyGroup Condition=" '$(Configuration)|$(Platform)' == 'Release|x86' ">
<PlatformTarget>x86</PlatformTarget>
<DebugType>pdbonly</DebugType>
<Optimize>true</Optimize>
<OutputPath>bin\Release\</OutputPath>
<DefineConstants>TRACE</DefineConstants>
<ErrorReport>prompt</ErrorReport>
<WarningLevel>4</WarningLevel>
</PropertyGroup>
<ItemGroup>
<Reference Include="Microsoft.Research.Kinect, Version=1.0.0.45, Culture=neutral, PublicKeyToken=31bf3856ad364e35, processorArchitecture=MSIL" />
<Reference Include="Microsoft.Speech, Version=10.2.0.0, Culture=neutral, PublicKeyToken=31bf3856ad364e35, processorArchitecture=MSIL">
<SpecificVersion>False</SpecificVersion>
<HintPath>C:\Program Files\Microsoft Speech Platform SDK\Assembly\Microsoft.Speech.dll</HintPath>
</Reference>
<Reference Include="System" />
<Reference Include="System.Core" />
<Reference Include="System.Xml.Linq" />
<Reference Include="System.Data.DataSetExtensions" />
<Reference Include="Microsoft.CSharp" />
<Reference Include="System.Data" />
<Reference Include="System.Xml" />
</ItemGroup>
<ItemGroup>
<Compile Include="Program.cs" />
<Compile Include="Properties\AssemblyInfo.cs" />
<Compile Include="SpeechToText.cs" />
</ItemGroup>
<Import Project="$(MSBuildToolsPath)\Microsoft.CSharp.targets" />
<!-- To modify your build process, add your task inside one of the targets below and uncomment it.
Other similar extension points exist, see Microsoft.Common.targets.
<Target Name="BeforeBuild">
</Target>
<Target Name="AfterBuild">
</Target>
-->
</Project>
+35
View File
@@ -0,0 +1,35 @@
using System;
using System.Collections.Generic;
using System.Linq;
using System.Text;
namespace Demo3
{
class Program
{
static void Main(string[] args)
{
KinectWorkshop.SpeechToText stt = new KinectWorkshop.SpeechToText();
stt.TextRecognized += new EventHandler<KinectWorkshop.SpeechToText.TextRecognizedArgs>(stt_TextRecognized);
stt.Start();
Console.ReadLine();
stt.Stop();
}
static void stt_TextRecognized(object sender, KinectWorkshop.SpeechToText.TextRecognizedArgs e)
{
Console.WriteLine(e.Verb);
switch (e.Verb)
{
case KinectWorkshop.SpeechToText.Verbs.None:
break;
case KinectWorkshop.SpeechToText.Verbs.PreviousSlide:
break;
case KinectWorkshop.SpeechToText.Verbs.NextSlide:
break;
default:
break;
}
}
}
}
@@ -0,0 +1,36 @@
using System.Reflection;
using System.Runtime.CompilerServices;
using System.Runtime.InteropServices;
// Allgemeine Informationen über eine Assembly werden über die folgenden
// Attribute gesteuert. Ändern Sie diese Attributwerte, um die Informationen zu ändern,
// die mit einer Assembly verknüpft sind.
[assembly: AssemblyTitle("Demo3")]
[assembly: AssemblyDescription("")]
[assembly: AssemblyConfiguration("")]
[assembly: AssemblyCompany("")]
[assembly: AssemblyProduct("Demo3")]
[assembly: AssemblyCopyright("Copyright © 2011")]
[assembly: AssemblyTrademark("")]
[assembly: AssemblyCulture("")]
// Durch Festlegen von ComVisible auf "false" werden die Typen in dieser Assembly unsichtbar
// für COM-Komponenten. Wenn Sie auf einen Typ in dieser Assembly von
// COM zugreifen müssen, legen Sie das ComVisible-Attribut für diesen Typ auf "true" fest.
[assembly: ComVisible(false)]
// Die folgende GUID bestimmt die ID der Typbibliothek, wenn dieses Projekt für COM verfügbar gemacht wird
[assembly: Guid("1fd608c9-3dd2-4d4b-b7a0-c368ff56a377")]
// Versionsinformationen für eine Assembly bestehen aus den folgenden vier Werten:
//
// Hauptversion
// Nebenversion
// Buildnummer
// Revision
//
// Sie können alle Werte angeben oder die standardmäßigen Build- und Revisionsnummern
// übernehmen, indem Sie "*" eingeben:
// [assembly: AssemblyVersion("1.0.*")]
[assembly: AssemblyVersion("1.0.0.0")]
[assembly: AssemblyFileVersion("1.0.0.0")]
+136
View File
@@ -0,0 +1,136 @@
using System;
using System.Collections.Generic;
using System.Diagnostics;
using System.IO;
using System.Linq;
using System.Threading;
using Microsoft.Research.Kinect.Audio;
using Microsoft.Speech.AudioFormat;
using Microsoft.Speech.Recognition;
namespace KinectWorkshop
{
public class SpeechToText
{
public enum Verbs
{
None = 0,
PreviousSlide,
NextSlide
};
Dictionary<string, Verbs> SinglePhrases = new Dictionary<string, Verbs>()
{
{"previous slide", Verbs.PreviousSlide},
{"next slide", Verbs.NextSlide}
};
public class TextRecognizedArgs : EventArgs
{
public Verbs Verb { get; set; }
public string Phrase { get; set; }
public string Matched {get; set; }
}
public event EventHandler<TextRecognizedArgs> TextRecognized;
private KinectAudioSource kinectSource;
private SpeechRecognitionEngine sre;
private const string RecognizerId = "SR_MS_en-US_Kinect_10.0";
public SpeechToText()
{
RecognizerInfo ri = SpeechRecognitionEngine.InstalledRecognizers().Where(r => r.Id == RecognizerId).FirstOrDefault();
if (ri == null)
return;
sre = new SpeechRecognitionEngine(ri.Id);
var singlePhrases = new Choices();
foreach (var phrase in SinglePhrases)
singlePhrases.Add(phrase.Key);
var g = new Grammar(new GrammarBuilder(singlePhrases));
sre.LoadGrammar(g);
sre.SpeechRecognized += sre_SpeechRecognized;
sre.SpeechRecognitionRejected += new EventHandler<SpeechRecognitionRejectedEventArgs>(sre_SpeechRecognitionRejected);
}
private void StartSST()
{
kinectSource = new KinectAudioSource();
kinectSource.SystemMode = SystemMode.OptibeamArrayOnly;
kinectSource.FeatureMode = true;
kinectSource.AutomaticGainControl = false;
kinectSource.MicArrayMode = MicArrayMode.MicArrayAdaptiveBeam;
var kinectStream = kinectSource.Start();
sre.SetInputToAudioStream(kinectStream, new SpeechAudioFormatInfo(
EncodingFormat.Pcm, 16000, 16, 1,
32000, 2, null));
sre.RecognizeAsync(RecognizeMode.Multiple);
//const double alpha = 0.35;
//double angle = 0.0;
//while (true)
//{
// if (kinectSource.SoundSourcePositionConfidence > 0.5)
// {
// double a = alpha*kinectSource.SoundSourcePositionConfidence;
// angle = (1 - a) * angle + a * kinectSource.SoundSourcePosition;
// Console.WriteLine(angle + ":" + kinectSource.MicArrayBeamAngle);
// }
// System.Threading.Thread.Sleep(50);
//}
}
public void Start()
{
var t = new Thread(StartSST);
t.Start();
}
public void Stop()
{
if (sre != null)
{
sre.RecognizeAsyncCancel();
sre.RecognizeAsyncStop();
kinectSource.Dispose();
}
}
void sre_SpeechRecognitionRejected(object sender, SpeechRecognitionRejectedEventArgs e)
{
var said = new TextRecognizedArgs();
said.Verb = Verbs.None;
said.Matched = "?";
TextRecognized(new object(), said);
Console.WriteLine("\nSpeech Rejected");
}
void sre_SpeechRecognized(object sender, SpeechRecognizedEventArgs e)
{
if (TextRecognized == null)
return;
var said = new TextRecognizedArgs();
said.Verb = Verbs.None;
said.Phrase = e.Result.Text;
bool found = false;
foreach (var phrase in SinglePhrases)
{
if (e.Result.Text.Contains(phrase.Key))
{
said.Verb = phrase.Value;
found = true;
break;
}
}
if (!found)
return;
TextRecognized(new object(), said);
}
}
}