private static void SynthToCam(string text, CameraWindow cw)
{
var synthFormat = new System.Speech.AudioFormat.SpeechAudioFormatInfo(System.Speech.AudioFormat.EncodingFormat.Pcm, 11025, 16, 1, 22100, 2, null);
using (var synthesizer = new SpeechSynthesizer())
{
using (var waveStream = new MemoryStream())
{
//write some silence to the stream to allow camera to initialise properly
var silence = new byte[1 * 22050];
waveStream.Write(silence, 0, silence.Count());
var pbuilder = new PromptBuilder();
var pStyle = new PromptStyle
{
Emphasis = PromptEmphasis.Strong,
Rate = PromptRate.Slow,
Volume = PromptVolume.ExtraLoud
};
pbuilder.StartStyle(pStyle);
pbuilder.StartParagraph();
pbuilder.StartVoice(VoiceGender.Male, VoiceAge.Adult, 2);
pbuilder.StartSentence();
pbuilder.AppendText(text);
pbuilder.EndSentence();
pbuilder.EndVoice();
pbuilder.EndParagraph();
pbuilder.EndStyle();
synthesizer.SetOutputToAudioStream(waveStream, synthFormat);
synthesizer.Speak(pbuilder);
synthesizer.SetOutputToNull();
//write some silence to the stream to allow camera to end properly
waveStream.Write(silence, 0, silence.Count());
waveStream.Seek(0, SeekOrigin.Begin);
var ds = new DirectStream(waveStream)
{
RecordingFormat = new WaveFormat(11025, 16, 1)
};
var talkTarget = TalkHelper.GetTalkTarget(cw.Camobject, ds);
ds.Start();
talkTarget.Start();
while (ds.IsRunning)
{
Thread.Sleep(100);
}
ds.Stop();
talkTarget.Stop();
talkTarget = null;
ds = null;
waveStream.Close();
}
}
}
private static void SynthToCam(string text, CameraWindow cw)
{
var synthFormat = new System.Speech.AudioFormat.SpeechAudioFormatInfo(System.Speech.AudioFormat.EncodingFormat.Pcm, 11025, 16, 1, 22100, 2, null);
using (var synthesizer = new SpeechSynthesizer())
{
using (var waveStream = new MemoryStream())
{
//write some silence to the stream to allow camera to initialise properly
var silence = new byte[1 * 22050];
waveStream.Write(silence, 0, silence.Length);
var pbuilder = new PromptBuilder();
var pStyle = new PromptStyle
{
Emphasis = PromptEmphasis.Strong,
Rate = PromptRate.Slow,
Volume = PromptVolume.ExtraLoud
};
pbuilder.StartStyle(pStyle);
pbuilder.StartParagraph();
pbuilder.StartVoice(VoiceGender.Male, VoiceAge.Adult, 2);
pbuilder.StartSentence();
pbuilder.AppendText(text);
pbuilder.EndSentence();
pbuilder.EndVoice();
pbuilder.EndParagraph();
pbuilder.EndStyle();
synthesizer.SetOutputToAudioStream(waveStream, synthFormat);
synthesizer.Speak(pbuilder);
synthesizer.SetOutputToNull();
//write some silence to the stream to allow camera to end properly
waveStream.Write(silence, 0, silence.Length);
waveStream.Seek(0, SeekOrigin.Begin);
var ds = new DirectStream(waveStream) { RecordingFormat = new WaveFormat(11025, 16, 1) };
var talkTarget = TalkHelper.GetTalkTarget(cw.Camobject, ds);
ds.Start();
talkTarget.Start();
while (ds.IsRunning)
{
Thread.Sleep(100);
}
ds.Stop();
talkTarget.Stop();
talkTarget = null;
ds = null;
}
}
}
private void makeWaveButton_Click(object sender, RoutedEventArgs e)
{
System.Speech.AudioFormat.SpeechAudioFormatInfo formatInfo = new System.Speech.AudioFormat.SpeechAudioFormatInfo(8000, System.Speech.AudioFormat.AudioBitsPerSample.Eight, System.Speech.AudioFormat.AudioChannel.Mono);
speaker.SetOutputToWaveFile(PathToWav, formatInfo);
speaker.Speak(textBox1.Text);
speaker.SetOutputToDefaultAudioDevice();
result = true;
speaker.Dispose();
Close();
}
public static async System.Threading.Tasks.Task <Byte[]> TextToSpeechServiceAsync(string text, System.Collections.Generic.Dictionary <string, string> apiArgs)
{
Log.WriteLine("text:\"" + text + "\"");
//System.Globalization.CultureInfo ci = new System.Globalization.CultureInfo(Options.options.locale.language);
using (System.Speech.Synthesis.SpeechSynthesizer synth = new System.Speech.Synthesis.SpeechSynthesizer())
{
// Explicitly specify audio settings. All services are ok with 16000/16/1. It's ok to cast options to enums as their values are identical.
int sampleRate = int.Parse(apiArgs["sampleRate"]);
System.Speech.AudioFormat.SpeechAudioFormatInfo si = new System.Speech.AudioFormat.SpeechAudioFormatInfo(sampleRate, (System.Speech.AudioFormat.AudioBitsPerSample)WoundifyShared.Options.options.audio.bitDepth, (System.Speech.AudioFormat.AudioChannel)WoundifyShared.Options.options.audio.channels);
// TODO: use memory based file instead
synth.SetOutputToWaveFile(Options.options.tempFolderPath + Options.options.audio.speechSynthesisFileName, si);
synth.SelectVoiceByHints((System.Speech.Synthesis.VoiceGender)Options.commandservices["TextToSpeech"].voiceGender, (System.Speech.Synthesis.VoiceAge)Options.commandservices["TextToSpeech"].voiceAge);
synth.Speak(text);
}
return(await Helpers.ReadBytesFromFileAsync(Options.options.audio.speechSynthesisFileName));
}
private void Save_BTN_Click(object sender, RoutedEventArgs e)
{
SetOutputPath();
synth.SelectVoice(installedVoices[VoiceDropdown.SelectedIndex].VoiceInfo.Name);
/////
///Set Speech Audio Formatting
///
System.Speech.AudioFormat.SpeechAudioFormatInfo audioFormatInfo = new System.Speech.AudioFormat.SpeechAudioFormatInfo(84300, System.Speech.AudioFormat.AudioBitsPerSample.Sixteen, System.Speech.AudioFormat.AudioChannel.Stereo);
synth.SetOutputToWaveFile(SetOutputPath(), audioFormatInfo);
synth.Speak(TextToSpeak.Text);
synth.SetOutputToNull();
}
public async Task Speak(string text, BufferedWaveProvider waveProvider, int rate)
{
var fmt = new System.Speech.AudioFormat.SpeechAudioFormatInfo(waveProvider.WaveFormat.SampleRate, (System.Speech.AudioFormat.AudioBitsPerSample)waveProvider.WaveFormat.BitsPerSample, (System.Speech.AudioFormat.AudioChannel)waveProvider.WaveFormat.Channels);
var _synthesizer = new SpeechSynthesizer()
{
Volume = 100, // 0...100
Rate = rate, // -10...10
};
_synthesizer.SelectVoice(Voice.VoiceInfo.Name);
_synthesizer.SetOutputToAudioStream(new OutputStream(waveProvider), fmt);
_synthesizer.SpeakAsync(text);
await Task.FromResult(0);
}
public async Task Speak(string text, BufferedWaveProvider waveProvider, int rate)
{
var fmt = new System.Speech.AudioFormat.SpeechAudioFormatInfo(waveProvider.WaveFormat.SampleRate, (System.Speech.AudioFormat.AudioBitsPerSample)waveProvider.WaveFormat.BitsPerSample, (System.Speech.AudioFormat.AudioChannel)waveProvider.WaveFormat.Channels);
// Creates an instance of a speech config with specified subscription key and service region.
// Replace with your own subscription key and service region (e.g., "westus").
var config = SpeechConfig.FromSubscription(Key, Region);
config.SpeechSynthesisLanguage = Language;
config.SpeechSynthesisVoiceName = Voice;
// Creates an audio out stream.
using (var stream = AudioOutputStream.CreatePullStream(AudioStreamFormat.GetWaveFormatPCM((uint)waveProvider.WaveFormat.SampleRate, (byte)waveProvider.WaveFormat.BitsPerSample, (byte)waveProvider.WaveFormat.Channels)))
{
// Creates a speech synthesizer using audio stream output.
using (var streamConfig = AudioConfig.FromStreamOutput(stream))
using (var synthesizer = new SpeechSynthesizer(config, streamConfig))
{
using (var result = await synthesizer.SpeakTextAsync(text))
{
if (result.Reason == ResultReason.SynthesizingAudioCompleted)
{
//Console.WriteLine($"Speech synthesized for text [{text}], and the audio was written to output stream.");
}
else if (result.Reason == ResultReason.Canceled)
{
var cancellation = SpeechSynthesisCancellationDetails.FromResult(result);
OnLog?.Invoke($"CANCELED: Reason={cancellation.Reason}");
if (cancellation.Reason == CancellationReason.Error)
{
OnLog?.Invoke($"CANCELED: ErrorCode={cancellation.ErrorCode}");
OnLog?.Invoke($"CANCELED: ErrorDetails=[{cancellation.ErrorDetails}]");
OnLog?.Invoke($"CANCELED: Did you update the subscription info?");
}
}
}
}
/*
* using (var reader = new WaveFileReader(new PullStream(stream)))
* {
* var newFormat = new WaveFormat(waveProvider.WaveFormat.SampleRate, waveProvider.WaveFormat.BitsPerSample, waveProvider.WaveFormat.Channels);
* using (var conversionStream = new WaveFormatConversionStream(newFormat, reader))
* {
* //WaveFileWriter.CreateWaveFile("output.wav", conversionStream);
* byte[] buffer = new byte[32000];
* int filledSize = 0;
* int totalSize = 0;
* while ((filledSize = conversionStream.Read(buffer, 0, buffer.Length)) > 0)
* {
* waveProvider.AddSamples(buffer, 0, (int)filledSize);
* //Console.WriteLine($"{filledSize} bytes received.");
* totalSize += filledSize;
* }
* }
* }*/
// Reads(pulls) data from the stream
byte[] buffer = new byte[32000];
uint filledSize = 0;
uint totalSize = 0;
while ((filledSize = stream.Read(buffer)) > 0)
{
waveProvider.AddSamples(buffer, 0, (int)filledSize);
//Console.WriteLine($"{filledSize} bytes received.");
totalSize += filledSize;
}
}
}
static void Main(string[] args)
{
SpeechSynthesizer s = new SpeechSynthesizer();
int Volume = s.Volume;
int Rate = s.Rate;
string line;
bool mode_interactive = false;
bool mode_tee = false;
bool mode_scroll_inspect = false;
bool mode_report_progress = false;
int consumed = 0; // the number of args that are used up - and thusly not the payload
string InputFilename = "";
string OutputFilename = "";
// sound properties
System.Speech.AudioFormat.AudioChannel numchannels = System.Speech.AudioFormat.AudioChannel.Stereo;
System.Speech.AudioFormat.AudioBitsPerSample numbits = System.Speech.AudioFormat.AudioBitsPerSample.Sixteen;
Int32 khz = 44;
// So... unicode. Notepad for instance can save accented characters in ANSI mode but you really want utf-8
bool unicode_warning_sent = false;
// This annoys me that I had to do this.
bool keyAvailableHack = false;
try
{
keyAvailableHack = System.Console.KeyAvailable;
}
catch
{
keyAvailableHack = true;
}
// help for no args unless we are piping
if (args.Length == 0 && keyAvailableHack == false)
{
Help();
Environment.Exit(0);
}
for (int i = 0; i < args.Length; i++)
{
if (args[i] == "-v" || args[i] == "--volume" || args[i] == "/v" || args[i] == "/volume")
{
try
{
Volume = Convert.ToInt32(args[i + 1]);
if (Volume < 0 || Volume > 100)
{
Console.WriteLine("Volume must be between 0 and 100, inclusive.");
Environment.Exit(1);
}
}
catch
{
Console.WriteLine("Invalid volume specified");
Environment.Exit(1);
}
i++;
consumed += 2;
}
if (args[i] == "-r" || args[i] == "--rate" || args[i] == "/r" || args[i] == "/rate")
{
try
{
Rate = Convert.ToInt32(args[i + 1]);
if (Rate < -10 || Rate > 10)
{
Console.WriteLine("Rate must be between -10 and 10, inclusive.");
Environment.Exit(1);
}
}
catch
{
Console.WriteLine("Invalid rate specified");
Environment.Exit(1);
}
i++;
consumed += 2;
}
if (args[i] == "-m" || args[i] == "--male" || args[i] == "/m" || args[i] == "/male")
{
s.SelectVoiceByHints(VoiceGender.Male);
consumed++;
}
if (args[i] == "-f" || args[i] == "--female" || args[i] == "/f" || args[i] == "/female")
{
s.SelectVoiceByHints(VoiceGender.Female);
consumed++;
}
if (args[i] == "-n" || args[i] == "--name" || args[i] == "/n" || args[i] == "/name")
{
try
{
s.SelectVoice(args[i + 1]);
}
catch
{
Console.WriteLine("Invalid voice name specified. Did you use quotes? Try using --list to see available names");
Environment.Exit(1);
}
i++;
consumed += 2;
}
if (args[i] == "-l" || args[i] == "--list" || args[i] == "/l" || args[i] == "/list")
{
//choose by name
foreach (InstalledVoice v in s.GetInstalledVoices())
{
Console.WriteLine("\"{0}\" - {1},{2},{3}", v.VoiceInfo.Name, v.VoiceInfo.Age, v.VoiceInfo.Gender, v.VoiceInfo.Culture);
//Console.WriteLine(v.VoiceInfo.Description);
//Console.WriteLine(v.VoiceInfo.Id);
}
Environment.Exit(0);
}
if (args[i] == "-?" || args[i] == "--?" || args[i] == "/?" || args[i] == "/h" || args[i] == "-h" || args[i] == "/help" || args[i] == "--help" || args[i] == "-help")
{
//help and version and quit
Help();
Environment.Exit(0);
}
if (args[i] == "-i" || args[i] == "--interactive" || args[i] == "/i" || args[i] == "/interactive")
{
mode_interactive = true;
consumed++;
}
if (args[i] == "-t" || args[i] == "--tee" || args[i] == "/t" || args[i] == "/tee")
{
mode_tee = true;
consumed++;
}
if (args[i] == "-s" || args[i] == "--scroll" || args[i] == "/s" || args[i] == "/scroll")
{
mode_scroll_inspect = true;
consumed++;
}
if (args[i] == "-p" || args[i] == "--progress" || args[i] == "/p" || args[i] == "/progress")
{
mode_report_progress = true;
consumed++;
}
if (args[i] == "-d" || args[i] == "--done" || args[i] == "/d" || args[i] == "/done")
{
// nothing after "done" gets parsed, in case there are actual codes in the speech
consumed++;
break;
}
if (args[i] == "-k" || args[i] == "--input" || args[i] == "/k" || args[i] == "/input")
{
InputFilename = args[i + 1];
i++;
consumed += 2;
}
if (args[i] == "-o" || args[i] == "--output" || args[i] == "/o" || args[i] == "/output")
{
OutputFilename = args[i + 1];
i++;
consumed += 2;
}
if (args[i] == "--mono" || args[i] == "/mono")
{
numchannels = System.Speech.AudioFormat.AudioChannel.Mono;
consumed++;
}
if (args[i] == "--8bit" || args[i] == "/8bit")
{
numbits = System.Speech.AudioFormat.AudioBitsPerSample.Eight;
consumed++;
}
if (args[i] == "--khz" || args[i] == "/khz")
{
khz = Convert.ToInt32(args[i + 1]);
i++;
consumed += 2;
}
}
if (mode_report_progress)
{
s.SpeakProgress += new EventHandler <SpeakProgressEventArgs>(report_progress);
}
s.Volume = Volume;
s.Rate = Rate;
System.Speech.AudioFormat.SpeechAudioFormatInfo fmt = new System.Speech.AudioFormat.SpeechAudioFormatInfo(khz * 1000, numbits, numchannels);
if (OutputFilename != "")
{
try
{
s.SetOutputToWaveFile(OutputFilename, fmt);
}
catch (Exception e)
{
Console.WriteLine("Error, the output file could not be written.");
Console.WriteLine(e.Message);
}
}
if (InputFilename != "")
{
try
{
using (StreamReader sr = new StreamReader(InputFilename, true))
{
while ((line = sr.ReadLine()) != null)
{
// Including this warning because it is apparently very easy (I did it, anyway) to save what you *think* are properly
// accented non-english characters in a notepad file under ANSI - they show up properly in notepad etc but when you
// feed them to Speech.Synthesis they are replaced with the 65533 garbage character.
if (unicode_warning_sent == false)
{
for (int i = 0; i < line.Length; i++)
{
if (Convert.ToInt32(line[i]) == 65533)
{
Console.WriteLine("Warning: Unicode replacement character 65533 detected. Check encoding on input file.");
unicode_warning_sent = true;
break;
}
}
}
s.Speak(line);
}
}
}
catch (Exception e)
{
Console.WriteLine("The input file could not be read:");
Console.WriteLine(e.Message);
}
}
// "from stdin, noninteractive" mode
if (keyAvailableHack)
{
while (true)
{
line = Console.ReadLine();
if (line == null)
{
break;
}
else
{
if (mode_scroll_inspect)
{
// check for scroll lock
if ((((ushort)GetKeyState(0x91)) & 0xffff) != 0)
{
Console.Error.WriteLine(line);
s.Speak(line);
}
Console.WriteLine(line);
continue;
}
if (mode_tee)
{
Console.WriteLine(line);
}
s.Speak(line);
}
}
Environment.Exit(0);
}
// from stdin, interactive
// "from stdin, noninteractive" mode
if (mode_interactive)
{
Console.WriteLine("Interactive mode. Control-c to quit.");
while (true)
{
line = Console.ReadLine();
if (line != null)
{
s.Speak(line);
}
else
{
break;
}
}
Environment.Exit(0);
}
// "from the args" mode
string tosay = String.Join(" ", args.Skip(consumed));
if (tosay.Length > 0)
{
s.Speak(tosay);
}
Environment.Exit(0);
}
private static void SynthToCam(string text, CameraWindow cw)
{
var synthFormat = new System.Speech.AudioFormat.SpeechAudioFormatInfo(System.Speech.AudioFormat.EncodingFormat.Pcm, 11025, 16, 1, 22100, 2, null);
using (var synthesizer = new SpeechSynthesizer())
{
using (var waveStream = new MemoryStream())
{
//write some silence to the stream to allow camera to initialise properly
var silence = new byte[1 * 22050];
waveStream.Write(silence, 0, silence.Count());
var pbuilder = new PromptBuilder();
var pStyle = new PromptStyle
{
Emphasis = PromptEmphasis.Strong,
Rate = PromptRate.Slow,
Volume = PromptVolume.ExtraLoud
};
pbuilder.StartStyle(pStyle);
pbuilder.StartParagraph();
pbuilder.StartVoice(VoiceGender.Male, VoiceAge.Adult, 2);
pbuilder.StartSentence();
pbuilder.AppendText(text);
pbuilder.EndSentence();
pbuilder.EndVoice();
pbuilder.EndParagraph();
pbuilder.EndStyle();
synthesizer.SetOutputToAudioStream(waveStream, synthFormat);
synthesizer.Speak(pbuilder);
synthesizer.SetOutputToNull();
//write some silence to the stream to allow camera to end properly
waveStream.Write(silence, 0, silence.Count());
waveStream.Seek(0, SeekOrigin.Begin);
ITalkTarget talkTarget = null;
var ds = new DirectStream(waveStream)
{
RecordingFormat = new WaveFormat(11025, 16, 1)
};
switch (cw.Camobject.settings.audiomodel)
{
case "Foscam":
ds.Interval = 40;
ds.PacketSize = 882; // (40ms packet at 22050 bytes per second)
talkTarget = new TalkFoscam(cw.Camobject.settings.audioip, cw.Camobject.settings.audioport,
cw.Camobject.settings.audiousername,
cw.Camobject.settings.audiopassword, ds);
break;
case "NetworkKinect":
ds.Interval = 40;
ds.PacketSize = 882;
talkTarget = new TalkNetworkKinect(cw.Camobject.settings.audioip, cw.Camobject.settings.audioport, ds);
break;
case "iSpyServer":
ds.Interval = 40;
ds.PacketSize = 882;
talkTarget = new TalkiSpyServer(cw.Camobject.settings.audioip,
cw.Camobject.settings.audioport,
ds);
break;
case "Axis":
talkTarget = new TalkAxis(cw.Camobject.settings.audioip, cw.Camobject.settings.audioport,
cw.Camobject.settings.audiousername,
cw.Camobject.settings.audiopassword, ds);
break;
default:
//local playback
talkTarget = new TalkLocal(ds);
break;
}
ds.Start();
talkTarget.Start();
while (ds.IsRunning)
{
Thread.Sleep(100);
}
ds.Stop();
if (talkTarget != null)
{
talkTarget.Stop();
}
talkTarget = null;
ds = null;
waveStream.Close();
}
}
}
