I have recently decided to look at C# because I found a cool tutorial on YouTube about creating a JARVIS speech recognition program in it. I have followed the tutorial completely and have managed to get it to run without any errors at all. However, it is not responding to my voice input, as can be seen in the code an input is "hello" I have tried that amongst others and can't get it to respond. According to the comments on the video I am not the only one it's happening to, could you please help me find what's causing this? thank you. Here's the full code:
using System;
using System.Collections.Generic;
using System.ComponentModel;
using System.Data;
using System.Drawing;
using System.Linq;
using System.Text;
using System.Threading.Tasks;
using System.Windows.Forms;
using System.Speech.Recognition;
using System.Speech.Synthesis;
using System.IO;
using System.Xml;
namespace J.A.R.V.I.S
{
public partial class Form1 : Form
{
SpeechRecognitionEngine _recognizer = new SpeechRecognitionEngine();
SpeechSynthesizer JARVIS = new SpeechSynthesizer();
string QEvent;
string ProcWindow;
double timer = 10;
int count = 1;
Random rnd = new Random();
public Form1()
{
InitializeComponent();
}
private void Form1_Load(object sender, EventArgs e)
{
_recognizer.SetInputToDefaultAudioDevice();
_recognizer.LoadGrammar(new DictationGrammar());
_recognizer.LoadGrammar(new Grammar(new GrammarBuilder(new Choices(File.ReadAllLines(@"C:\Users\chaye\Documents\Commands.txt")))));
_recognizer.SpeechRecognized += new EventHandler<SpeechRecognizedEventArgs>(_recognizer_SpeechRecognized);
_recognizer.RecognizeAsync(RecognizeMode.Multiple);
}
void _recognizer_SpeechRecognized(object sender, SpeechRecognizedEventArgs e)
{
int ranNum = rnd.Next(1, 10);
string speech = e.Result.Text;
switch (speech)
{
//GREETINGS
case "hello":
case "hello jarvis":
if (ranNum < 6) { JARVIS.Speak("Hello sir"); }
else if (ranNum > 5) { JARVIS.Speak("Hi"); }
break;
case "goodbye":
case "goodbye jarvis":
case "close":
case "close jarvis":
JARVIS.Speak("Until next time");
Close();
break;
case "jarvis":
if (ranNum < 5) { QEvent = ""; JARVIS.Speak("Yes sir"); }
else if (ranNum > 4) { QEvent = ""; JARVIS.Speak("Yes?"); }
break;
//WEBSITES
case "open youtube":
System.Diagnostics.Process.Start("http://www.youtube.com");
break;
case "open google":
System.Diagnostics.Process.Start("http://www.google.com");
break;
//SHELL COMMANDS
case "open program":
System.Diagnostics.Process.Start("file location");
JARVIS.Speak("Loading");
break;
case "open Word":
System.Diagnostics.Process.Start(@"C:\Program Files\Microsoft Office\Office15\WINWORD.EXE");
JARVIS.Speak("Loading");
break;
//CLOSE PROGRAMS
case "close program":
ProcWindow = "process name";
StopWindow();
break;
case "close Skype":
ProcWindow = "Skype";
StopWindow();
break;
//CONDITION OF DAY
case "what time is it":
DateTime now = DateTime.Now;
string time = now.GetDateTimeFormats('t')[0];
JARVIS.Speak(time);
break;
case "what day is it":
JARVIS.Speak(DateTime.Today.ToString("dddd"));
break;
case "whats the date":
case "whats todays date":
JARVIS.Speak(DateTime.Today.ToString("dd-MM-yyyy"));
break;
//OTHER COMMANDS
case "go fullscreen":
FormBorderStyle = FormBorderStyle.None;
WindowState = FormWindowState.Maximized;
TopMost = true;
JARVIS.Speak("expanding");
break;
case "exit fullscreen":
FormBorderStyle = FormBorderStyle.Sizable;
WindowState = FormWindowState.Normal;
TopMost = false;
break;
case "switch window":
SendKeys.Send("%{TAB " + count + "}");
count += 1;
break;
case "reset":
count = 1;
timer = 11;
lblTimer.Visible = false;
ShutdownTimer.Enabled = false;
lstCommands.Visible = false;
break;
case "out of the way":
if (WindowState == FormWindowState.Normal || WindowState == FormWindowState.Maximized)
{
WindowState = FormWindowState.Minimized;
JARVIS.Speak("My apologies");
}
break;
case "come back":
if (WindowState == FormWindowState.Minimized)
{
JARVIS.Speak("Alright?");
WindowState = FormWindowState.Normal;
}
break;
case "show commands":
string[] commands = (File.ReadAllLines(@"txt file location"));
JARVIS.Speak("Very well");
lstCommands.Items.Clear();
lstCommands.SelectionMode = SelectionMode.None;
lstCommands.Visible = true;
foreach (string command in commands)
{
lstCommands.Items.Add(command);
}
break;
case "hide listbox":
lstCommands.Visible = false;
break;
//SHUTDOWN RESTART LOG OFF
case "shutdown":
if (ShutdownTimer.Enabled == false)
{
QEvent = "shutdown";
JARVIS.Speak("I will shutdown shortly");
lblTimer.Visible = true;
ShutdownTimer.Enabled = true;
}
break;
case "log off":
if (ShutdownTimer.Enabled == false)
{
QEvent = "logoff";
JARVIS.Speak("Logging off");
lblTimer.Visible = true;
ShutdownTimer.Enabled = true;
}
break;
case "restart":
if (ShutdownTimer.Enabled == false)
{
QEvent = "restart";
JARVIS.Speak("I'll be back shortly");
lblTimer.Visible = true;
ShutdownTimer.Enabled = true;
}
break;
case "abort":
if (ShutdownTimer.Enabled == true)
{
QEvent = "abort";
}
break;
case "speed up":
if (ShutdownTimer.Enabled == true)
{
ShutdownTimer.Interval = ShutdownTimer.Interval / 10;
}
break;
case "slow down":
if (ShutdownTimer.Enabled == true)
{
ShutdownTimer.Interval = ShutdownTimer.Interval * 10;
}
break;
}
}
private void ShutdownTimer_Tick(object sender, EventArgs e)
{
if (timer == 0)
{
lblTimer.Visible = false;
ComputerTermination();
ShutdownTimer.Enabled = false;
}
else if (QEvent == "abort")
{
timer = 10;
lblTimer.Visible = false;
ShutdownTimer.Enabled = false;
}
else
{
timer = timer - .01;
lblTimer.Text = timer.ToString();
}
}
private void ComputerTermination()
{
switch (QEvent)
{
case "shutdown":
System.Diagnostics.Process.Start("shutdown", "-s");
break;
case "logoff":
System.Diagnostics.Process.Start("shutdown", "-l");
break;
case "restart":
System.Diagnostics.Process.Start("shutdown", "-r");
break;
}
}
private void StopWindow()
{
System.Diagnostics.Process[] procs = System.Diagnostics.Process.GetProcessesByName(ProcWindow);
foreach (System.Diagnostics.Process proc in procs)
{
proc.CloseMainWindow();
}
}
}
}
I have also been working on speech recognition in c#. C# speech recognition doesn't work as you think. It matches best combination of obtained inputs and then proceeds according to that, so you have to code according to that. Best way is to use e.Result.Text to find what it actually recognizes when you say a word and then code that in like this.
using System;
using System.Collections.Generic;
using System.ComponentModel;
using System.Data;
using System.Drawing;
using System.Linq;
using System.Text;
using System.Threading.Tasks;
using System.Windows.Forms;
using System.Speech.Recognition;
using System.Speech.Synthesis;
using System.IO;
using System.Xml;
namespace J.A.R.V.I.S
{
public partial class Form1 : Form
{
SpeechRecognitionEngine _recognizer = new SpeechRecognitionEngine();
SpeechSynthesizer JARVIS = new SpeechSynthesizer();
string QEvent;
string ProcWindow;
double timer = 10;
int count = 1;
public Form1()
{
InitializeComponent();
}
private void Form1_Load(object sender, EventArgs e)
{
_recognizer.SetInputToDefaultAudioDevice();
_recognizer.LoadGrammar(new DictationGrammar());
_recognizer.LoadGrammar(new Grammar(new GrammarBuilder(new Choices(File.ReadAllLines(@"C:\Users\chaye\Documents\Commands.txt")))));
_recognizer.SpeechRecognized += new EventHandler<SpeechRecognizedEventArgs>(_recognizer_SpeechRecognized);
_recognizer.RecognizeAsync(RecognizeMode.Multiple);
}
void _recognizer_SpeechRecognized(object sender, SpeechRecognizedEventArgs e)
{
string speech = e.Result.Text;
MessageBox.Show(speech);
string[] start_array={"start","some other words that sounds same as start or result from e.Result.Text when you say start"};
if(start_array.Contains(speech))
{
SendKeys.SendWait("^{ESC}");
}
}
}
}