private void SpeakMessage(TextToSpeechMessage message)
{
var enabled = GlobalSettingsStore.Instance.ProfileSettingsStore.GetClientSettingBool(ProfileSettingsKeys.EnableTextToSpeech);
if (!enabled)
{
return;
}
Task task = new Task(() =>
{
try
{
using (var synth = new SpeechSynthesizer())
using (var stream = new MemoryStream())
{
synth.SelectVoiceByHints(VoiceGender.Female, VoiceAge.Adult, 0, new CultureInfo("en-GB", false));
synth.Rate = 1;
var vol = double.Parse(GlobalSettingsStore.Instance.ProfileSettingsStore.GetClientSetting(ProfileSettingsKeys.TextToSpeechVolume).RawValue, CultureInfo.InvariantCulture);
var intVol = (int)(vol * 100.0);
if (intVol > 100)
{
intVol = 100;
}
synth.Volume = intVol;
if (mixer.WaveFormat.Channels == 2)
{
synth.SetOutputToAudioStream(stream,
new SpeechAudioFormatInfo(mixer.WaveFormat.SampleRate, AudioBitsPerSample.Sixteen, AudioChannel.Stereo));
}
else
{
synth.SetOutputToAudioStream(stream,
new SpeechAudioFormatInfo(mixer.WaveFormat.SampleRate, AudioBitsPerSample.Sixteen, AudioChannel.Mono));
}
synth.Speak(message.Message);
//clear current message
buffer.ClearBuffer();
byte[] byteArr = stream.ToArray();
buffer.AddSamples(byteArr, 0, byteArr.Length);
}
}
catch (Exception ex)
{
Logger.Error(ex, "Error with Text to Speech");
}
});
task.Start();
}
public SpeechAnimator(ChannelSystem channelSystem, BoneSystem boneSystem)
{
this.channelSystem = channelSystem;
this.boneSystem = boneSystem;
this.synth = new SpeechSynthesizer();
synth.Volume = 100;
synth.SelectVoice(Voice);
synth.VisemeReached += Synth_VisemeReached;
visemeChannels = VisemeChannelMap
.Select(name => name == null ? null : channelSystem.ChannelsByName[name + "?value"])
.ToArray();
headBone = boneSystem.BonesByName["head"];
var audioDevice = new XAudio2(XAudio2Flags.DebugEngine, ProcessorSpecifier.AnyProcessor);
masteringVoice = new MasteringVoice(audioDevice);
WaveFormat monoWaveFormat = WaveFormat.CreateIeeeFloatWaveFormat(44100, 2);
sourceVoice = new SourceVoice(audioDevice, monoWaveFormat, VoiceFlags.None);
sourceVoice.SetVolume(1);
sourceVoice.Start();
phononStream = new PhononSourceVoiceStream(sourceVoice);
synth.SetOutputToAudioStream(phononStream, new SpeechAudioFormatInfo(44100, AudioBitsPerSample.Sixteen, AudioChannel.Mono));
}
private void speak(string text)
{
SpeechSynthesizer tts = new SpeechSynthesizer();
tts.SelectVoice((string)cmbTTS.SelectedItem);
tts.Volume = sliderTTSVol.Value * 5;
tts.Rate = sliderTTSSpeed.Value - 10;
MemoryStream ms = new MemoryStream();
tts.SetOutputToAudioStream(ms, formatInfo);
if (voiceStreams.Count == 0)
{
initVoiceStreams();
}
var toDiscordStream = safeGetVoiceStream();
tts.SpeakAsync(text);
tts.SpeakCompleted += async(a, b) => {
activeStreams.Add(toDiscordStream);
ms.Seek(0, SeekOrigin.Begin);
ms.CopyTo(toDiscordStream);
await toDiscordStream.FlushAsync();
activeStreams.Remove(toDiscordStream);
};
}
public void SpeakMessage(AudioVideoFlow flow, string message)
{
try
{
SpeechSynthesizer synth = new SpeechSynthesizer();
SpeechAudioFormatInfo formatInfo = new SpeechAudioFormatInfo(16000, AudioBitsPerSample.Sixteen, Microsoft.Speech.AudioFormat.AudioChannel.Mono);
SpeechSynthesisConnector connector = new SpeechSynthesisConnector();
synth.SetOutputToAudioStream(connector.Stream, formatInfo);
connector.AttachFlow(flow);
connector.Start();
synth.SpeakCompleted += new EventHandler<SpeakCompletedEventArgs>(
(sender, args) =>
{
connector.Stop();
synth.Dispose();
});
synth.SpeakAsync(message);
}
catch (Exception ex)
{
Console.WriteLine("Failed to play the message. {0}", ex);
}
}
private void speak(string text)
{
var delay = (1) * text.Split().Length;
while (DateTime.Now - whenSpeakRan < TimeSpan.FromSeconds(delay))
{
Thread.Sleep(125);
}
SpeechSynthesizer tts = new SpeechSynthesizer();
tts.SelectVoice(( string )cmbTTS.SelectedItem);
tts.Volume = sliderTTSVol.Value * 5;
tts.Rate = sliderTTSSpeed.Value - 10;
MemoryStream ms = new MemoryStream();
tts.SetOutputToAudioStream(ms, formatInfo);
if (voiceStream == null)
{
voiceStream = audioClient.CreatePCMStream(AudioApplication.Voice, 128 * 1024);
}
tts.SpeakAsync(text);
tts.SpeakCompleted += (a, b) => {
ms.Seek(0, SeekOrigin.Begin);
ms.CopyTo(voiceStream);
voiceStream.Flush();
};
whenSpeakRan = DateTime.Now;
}
/// <summary>
/// Convert XML to WAV bytes. WAV won't have the header, so you have to add it separately.
/// </summary>
static byte[] ConvertSsmlXmlToWav(string voiceId, string xml, WaveFormat format)
{
using (var ms = new MemoryStream())
{
using (var synthesizer = new SpeechSynthesizer())
{
//var format = new SpeechAudioFormatInfo(
if (format != null)
{
//var bps = format.BitsPerSample == 8 ? AudioBitsPerSample.Eight : AudioBitsPerSample.Sixteen;
var blockAlignment = format.BitsPerSample / 8 * format.Channels;
var averagerBytesPerSecond = format.SampleRate * format.BitsPerSample / 8 * format.Channels;
var formatInfo = new SpeechAudioFormatInfo(EncodingFormat.Pcm, format.SampleRate, format.BitsPerSample, format.Channels, averagerBytesPerSecond, blockAlignment, new byte[0]);
// Returns WAV data only.
synthesizer.SetOutputToAudioStream(ms, formatInfo);
}
try
{
var voice = synthesizer.GetInstalledVoices().Cast <InstalledVoice>().FirstOrDefault(x => x.VoiceInfo.Id == voiceId);
synthesizer.SelectVoice(voice.VoiceInfo.Name);
synthesizer.SpeakSsml(xml);
return(ms.ToArray());
}
catch (Exception ex)
{
ex.Data.Add("Voice", "voiceName");
OnEvent(Exception, ex);
}
}
}
return(null);
}
private byte[] TextToSpeech()
{
try
{
using (var synth = new SpeechSynthesizer())
using (var stream = new MemoryStream())
{
synth.SelectVoiceByHints(VoiceGender.Female, VoiceAge.Adult, 0, new CultureInfo("en-GB", false));
synth.Rate = 1;
var intVol = (int)(volume * 100.0);
if (intVol > 100)
{
intVol = 100;
}
synth.Volume = intVol;
synth.SetOutputToAudioStream(stream,
new SpeechAudioFormatInfo(INPUT_SAMPLE_RATE, AudioBitsPerSample.Sixteen, AudioChannel.Mono));
synth.Speak(path);
return(stream.ToArray());
}
}
catch (Exception ex)
{
Logger.Error(ex, "Error with Text to Speech");
}
return(new byte[0]);
}
public void Run()
{
// Create AudioVideoFlow
AudioVideoFlowHelper audioVideoFlowHelper = new AudioVideoFlowHelper();
_audioVideoFlow = audioVideoFlowHelper.CreateAudioVideoFlow(
null,
audioVideoFlow_StateChanged);
// Create a speech synthesis connector and attach it to an AudioVideoFlow
SpeechSynthesisConnector speechSynthesisConnector = new SpeechSynthesisConnector();
speechSynthesisConnector.AttachFlow(_audioVideoFlow);
// Create a speech synthesis and set connector to it
SpeechSynthesizer speechSynthesis = new SpeechSynthesizer();
SpeechAudioFormatInfo audioformat = new SpeechAudioFormatInfo(16000, AudioBitsPerSample.Sixteen, Microsoft.Speech.AudioFormat.AudioChannel.Mono);
speechSynthesis.SetOutputToAudioStream(speechSynthesisConnector, audioformat);
//Load readme file as the source
Console.WriteLine();
Console.Write("Please enter the source file => ");
string filename = Console.ReadLine();
string msg = "";
try
{
StreamReader objReader = new StreamReader(filename);
msg = objReader.ReadToEnd();
}
catch (FileNotFoundException)
{
Console.WriteLine("\r\nFile doesn't exist.");
ShutdownPlatform();
}
//Start connector
speechSynthesisConnector.Start();
Console.WriteLine("\r\nStreaming source file for 15 seconds.");
//Start streaming from speech synthesis.
speechSynthesis.SpeakAsync(new Prompt(msg));
//Allow the connector to stream 15 seconds by waiting for 15 seconds
Thread.Sleep(15000);
//Stop the connector
speechSynthesisConnector.Stop();
Console.WriteLine("\r\nSpeech synthesis connector stopped.");
//speech synthesis connector must be detached from the flow, otherwise if the connector is rooted, it will keep the flow in memory.
speechSynthesisConnector.DetachFlow();
// Shutdown the platform
ShutdownPlatform();
_waitForShutdownEventCompleted.WaitOne();
}
public void SynthesizeText(string text,
List <AudioMediaBuffer> audioMediaBuffers,
List <VideoMediaBuffer> videoMediaBuffers)
{
// stream for the output audio
var audioStream = new MemoryStream();
// set the synthesizer output to the stream, make sure the output format is matching the audio socket settings
_synth.SetOutputToAudioStream(audioStream,
new SpeechAudioFormatInfo(samplesPerSecond: 16000, bitsPerSample: AudioBitsPerSample.Sixteen,
channel: AudioChannel.Mono));
// observe the synthesizer to generate the visemes timeline
VisemesTimeline timeline = new VisemesTimeline();
_synth.VisemeReached += (sender, visemeReachedEventArgs) =>
{
timeline.Add(visemeReachedEventArgs.Viseme, visemeReachedEventArgs.Duration.Milliseconds);
};
// synthesize the text -> audio and visemes are generated
_synth.Speak(text);
// geneate the buffers and synchronize them with the current time
long referenceTimeTick = DateTime.Now.Ticks;
CreateAudioBuffers(audioStream, audioMediaBuffers, referenceTimeTick);
CreateVideoBuffers(timeline, videoMediaBuffers, referenceTimeTick);
}
public static string GetBase64Audio(string textInput)
{
var speechAudioFormatConfig = new SpeechAudioFormatInfo(samplesPerSecond: 8000, bitsPerSample: AudioBitsPerSample.Sixteen, channel: AudioChannel.Stereo);
var waveFormat = new WaveFormat(speechAudioFormatConfig.SamplesPerSecond, speechAudioFormatConfig.BitsPerSample, speechAudioFormatConfig.ChannelCount);
var prompt = new PromptBuilder
{
Culture = CultureInfo.CreateSpecificCulture("en-US")
};
prompt.StartVoice(prompt.Culture);
prompt.StartSentence();
prompt.StartStyle(new PromptStyle()
{
Emphasis = PromptEmphasis.Reduced,
Rate = PromptRate.Slow
});
prompt.AppendText(textInput);
prompt.EndStyle();
prompt.EndSentence();
prompt.EndVoice();
var mp3Stream = new MemoryStream();
byte[] audioOutputBytes;
string audioOutputAsString = null;
using (var synthWaveMemoryStream = new MemoryStream())
{
var resetEvent = new ManualResetEvent(false);
ThreadPool.QueueUserWorkItem(arg =>
{
try
{
var siteSpeechSynth = new SpeechSynthesizer();
siteSpeechSynth.SetOutputToAudioStream(synthWaveMemoryStream, speechAudioFormatConfig);
siteSpeechSynth.Speak(prompt);
}
finally
{
resetEvent.Set();
}
});
WaitHandle.WaitAll(new WaitHandle[] { resetEvent });
var bitRate = (speechAudioFormatConfig.AverageBytesPerSecond * 8);
synthWaveMemoryStream.Position = 0;
using (var mp3FileWriter = new LameMP3FileWriter(outStream: mp3Stream, format: waveFormat, bitRate: bitRate))
{
synthWaveMemoryStream.CopyTo(mp3FileWriter);
}
audioOutputBytes = mp3Stream.ToArray();
audioOutputAsString = $"data:audio/mp3;base64,{Convert.ToBase64String(audioOutputBytes)}";
}
return(audioOutputAsString);
}
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();
}
}
}
public virtual void PrepareMessageInStream(string message)
{
Stream = new MemoryStream();
_speechSynthesizer.SetOutputToAudioStream(Stream, _speechAudioFormatInfo);
_speechSynthesizer.Speak(message);
Stream.Position = 0;
}
public static TimeSpan NarrateText(string text, Stream outputStream)
{
var narrator = new SpeechSynthesizer();
narrator.SetOutputToAudioStream(outputStream, SpeechFormat);
long before = outputStream.Position;
narrator.Speak(text);
return(TimeSpan.FromSeconds((double)(outputStream.Position - before) / SpeechFormat.AverageBytesPerSecond));
}
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 SpeechSynthesizer CreateSpeechSynthesizer()
{
var synthesizer = new SpeechSynthesizer();
if (!string.IsNullOrEmpty(this.Configuration.Voice))
{
synthesizer.SelectVoice(this.Configuration.Voice);
}
if (this.Configuration.UseDefaultAudioPlaybackDevice)
{
// If specified, don't create an output stream and just set up the
// synthesizer to play sound directly to the default audio device.
synthesizer.SetOutputToDefaultAudioDevice();
}
else
{
// Create the format info from the configuration input format
SpeechAudioFormatInfo formatInfo = new SpeechAudioFormatInfo(
(EncodingFormat)this.Configuration.OutputFormat.FormatTag,
(int)this.Configuration.OutputFormat.SamplesPerSec,
this.Configuration.OutputFormat.BitsPerSample,
this.Configuration.OutputFormat.Channels,
(int)this.Configuration.OutputFormat.AvgBytesPerSec,
this.Configuration.OutputFormat.BlockAlign,
(this.Configuration.OutputFormat is WaveFormatEx) ? ((WaveFormatEx)this.Configuration.OutputFormat).ExtraInfo : null);
// Configure synthesizer to write to the output stream
synthesizer.SetOutputToAudioStream(
new IOStream(
(buffer, offset, count) =>
{
byte[] audioData = buffer;
if (buffer.Length != count)
{
audioData = new byte[count];
Array.Copy(buffer, offset, audioData, 0, count);
}
this.Out.Post(new AudioBuffer(audioData, this.Configuration.OutputFormat), this.pipeline.GetCurrentTime());
}),
formatInfo);
}
// Register all handlers
synthesizer.BookmarkReached += this.OnBookmarkReached;
synthesizer.PhonemeReached += this.OnPhonemeReached;
synthesizer.SpeakCompleted += this.OnSpeakCompleted;
synthesizer.SpeakProgress += this.OnSpeakProgress;
synthesizer.SpeakStarted += this.OnSpeakStarted;
synthesizer.StateChanged += this.OnStateChanged;
synthesizer.VisemeReached += this.OnVisemeReached;
return(synthesizer);
}
public void AddEndpoint(string ID, Stream outstream)
{
SpeechSynthesizer voice = new SpeechSynthesizer();
if (outstream == null) voice.SetOutputToDefaultAudioDevice();
else voice.SetOutputToAudioStream(outstream, new System.Speech.AudioFormat.SpeechAudioFormatInfo(
16000, System.Speech.AudioFormat.AudioBitsPerSample.Sixteen, System.Speech.AudioFormat.AudioChannel.Mono));
//if (chkIVONA.Checked) voice.SelectVoice("IVONA 2 Amy");
//else voice.SelectVoice("Microsoft Anna")
voices.Add(ID, voice);
outStreams.Add(ID, outstream);
}
public MemoryStream SynthesizeText(string text)
{
var audioStream = new MemoryStream();
// set the synthesizer output to the stream, make sure the output format is matching the audio socket settings
_synth.SetOutputToAudioStream(audioStream,
new SpeechAudioFormatInfo(samplesPerSecond: 16000, bitsPerSample: AudioBitsPerSample.Sixteen, channel: AudioChannel.Mono));
_synth.Speak(text);
return(audioStream);
}
public MemoryStream BeginRead(string toRead)
{
MemoryStream Bufferers = new MemoryStream();
Reader.SetOutputToAudioStream(Bufferers, new SpeechAudioFormatInfo(43500, AudioBitsPerSample.Sixteen, AudioChannel.Stereo));
if (toRead != null && toRead.Trim(' ') != "")
{ // await for a stream
Reader.Speak(toRead);
}
return(Bufferers);
}
private static void TextToSpeech(SpeechSynthesizer synth, string text)
{
var stream = new MemoryStream();
var format = new SpeechAudioFormatInfo(16000, AudioBitsPerSample.Sixteen, AudioChannel.Mono);
synth.SetOutputToAudioStream(stream, format);
synth.SpeakSsml(text);
var stdout = Console.OpenStandardOutput();
stream.Position = 0;
stream.CopyTo(stdout);
}
private async void Discord_ChatMessage(object sender, MessageEventArgs e)
{
if (this.beam.IsConnected && e.User.Id != this.discord.CurrentUser.Id && e.Channel.Id == 187760090206437376)
{
var msg = $"[discord] {e.Message.User.Name}: {e.Message.Text}";
await this.beam.SendMessage(msg);
}
if (this.readingMessages && e.Channel == this.readingChannel)
{
using (var synthesizer = new SpeechSynthesizer())
using (var mem = new MemoryStream())
{
var info = new SpeechAudioFormatInfo(48000, AudioBitsPerSample.Sixteen, AudioChannel.Stereo);
synthesizer.SetOutputToAudioStream(mem, info);
PromptBuilder builder = new PromptBuilder();
builder.Culture = CultureInfo.CreateSpecificCulture("en-US");
builder.StartVoice(builder.Culture);
builder.StartSentence();
builder.StartStyle(new PromptStyle()
{
Emphasis = PromptEmphasis.Reduced
});
builder.AppendText(e.Message.User.Name);
builder.AppendText(" says ");
builder.EndStyle();
builder.AppendText(e.Message.Text);
builder.EndSentence();
builder.EndVoice();
synthesizer.Speak(builder);
mem.Seek(0, SeekOrigin.Begin);
int count, block = 96000;
var buffer = new byte[block];
while ((count = mem.Read(buffer, 0, block)) > 0)
{
if (count < block)
{
for (int i = count; i < block; i++)
{
buffer[i] = 0;
}
}
this.audio.Send(buffer, 0, block);
}
}
}
}
public async Task <bool> PlayMessage(string message)
{
if (Util.IsUiThread)
{
return(await Task.Run(() => this.PlayMessage( message )));
}
await this.m_playbackSemaphore.WaitAsync();
Stream mStream = null;
try
{
if (this.IsPlaying)
{
return(false);
}
mStream = new MemoryStream(OneMeg);
var name = this.VoiceName;
using (var tts = new SpeechSynthesizer())
{
if (!string.IsNullOrEmpty(name))
{
tts.SelectVoice(name);
}
tts.SetOutputToAudioStream(mStream,
new SpeechAudioFormatInfo(44100, AudioBitsPerSample.Sixteen, AudioChannel.Stereo));
tts.Speak(message);
}
mStream.Flush();
mStream.Seek(0, SeekOrigin.Begin);
this.m_mixer.AddInputStream(new DisposingWaveProvider(mStream));
this.m_waveOut.Play();
}
catch (Exception e)
{
SimpleLogger.Log(nameof(SimpleAudioPlayer), $"Exception {e.GetType().Name} has occurred: {e.Message}");
mStream?.Dispose();
return(false);
}
finally
{
this.m_playbackSemaphore.Release();
}
return(true);
}
private byte[] LocalTTS(string msg)
{
try
{
using (var synth = new SpeechSynthesizer())
using (var stream = new MemoryStream())
{
if (opts.Voice == null || opts.Voice.Length == 0)
{
if (opts.Culture == null)
{
synth.SelectVoiceByHints(this.SpeakerGender, VoiceAge.Adult);
}
else
{
synth.SelectVoiceByHints(this.SpeakerGender, VoiceAge.Adult, 0, new CultureInfo(opts.Culture, false));
}
}
else
{
synth.SelectVoice(opts.Voice);
}
synth.Rate = opts.speed;
var intVol = (int)(opts.Volume * 100.0);
if (intVol > 100)
{
intVol = 100;
}
synth.Volume = intVol;
synth.SetOutputToAudioStream(stream,
new SpeechAudioFormatInfo(INPUT_SAMPLE_RATE, AudioBitsPerSample.Sixteen, AudioChannel.Mono));
synth.Speak(msg);
return(stream.ToArray());
}
}
catch (Exception ex)
{
Logger.Error(ex, $"Error with Microsoft Text to Speech: {ex.Message}");
}
return(new byte[0]);
}
public void AddSpeak(string text)
{
using (var stream = new MemoryStream())
{
_speaker.SetOutputToAudioStream(stream, new SpeechAudioFormatInfo(_config.SampleRate, AudioBitsPerSample.Sixteen, AudioChannel.Stereo));
_speaker.Speak(text);
_speaker.SetOutputToNull();
_writer.Write(stream.GetBuffer(), 0, stream.GetBuffer().Length);
AddEmpty(_config.WaitTime);
_writer.Write(stream.GetBuffer(), 0, stream.GetBuffer().Length);
AddEmpty(_config.WaitTime);
stream.Dispose();
}
}
private byte[] TextToSpeech()
{
try
{
using (var synth = new SpeechSynthesizer())
using (var stream = new MemoryStream())
{
bool isVoiceInstalled = false;
foreach (var voice in synth.GetInstalledVoices())
{
var info = voice.VoiceInfo;
if (this.SpeakerCulture == info.Culture.ToString())
{
isVoiceInstalled = true;
break;
}
}
if (!isVoiceInstalled)
{
this.SpeakerCulture = "en-GB";
}
synth.SelectVoiceByHints(this.SpeakerGender, VoiceAge.Adult, 0, new CultureInfo(this.SpeakerCulture, false));
synth.Rate = 1;
var intVol = (int)(volume * 100.0);
if (intVol > 100)
{
intVol = 100;
}
synth.Volume = intVol;
synth.SetOutputToAudioStream(stream,
new SpeechAudioFormatInfo(INPUT_SAMPLE_RATE, AudioBitsPerSample.Sixteen, AudioChannel.Mono));
synth.Speak(path);
return(stream.ToArray());
}
}
catch (Exception ex)
{
Logger.Error(ex, "Error with Text to Speech");
}
return(new byte[0]);
}
public Sync(SyncOp pOp)
{
AudioStream = new MemoryStream();
Synth = new SpeechSynthesizer();
ReloadLexicons();
Synth.SetOutputToAudioStream(AudioStream, new SpeechAudioFormatInfo(44100, AudioBitsPerSample.Sixteen, AudioChannel.Stereo)
{
});
_voices = Synth.GetInstalledVoices();
voiceArray = _voices?.Select(s => s.VoiceInfo.Name).ToArray();
_op = pOp;
_op(this, true);
}
public static async Task Say(VoiceNextConnection vnc, string say)
{
await vnc.SendSpeakingAsync(true); // send a speaking indicator
using (MemoryStream stream = new MemoryStream())
{
var info = new SpeechAudioFormatInfo(48000, AudioBitsPerSample.Sixteen, AudioChannel.Stereo);
using (SpeechSynthesizer synth = new SpeechSynthesizer())
{
//synth.SetOutputToAudioStream(stream, info);
synth.SetOutputToAudioStream(stream, info);
//var t = synth.GetInstalledVoices();
//synth.SelectVoice(t.First().VoiceInfo.Name);
synth.Speak(say);
synth.SetOutputToNull();
}
//await vnc.SendAsync(stream.ToArray(), 20, info.BitsPerSample);
stream.Seek(0, SeekOrigin.Begin);
Console.WriteLine("Format: {0}", info.EncodingFormat);
Console.WriteLine("BitRate: {0}", info.BitsPerSample);
Console.WriteLine("Block Alignment: {0}", info.BlockAlign);
Console.WriteLine("Samples per second: {0}", info.SamplesPerSecond);
var buff = new byte[3840];
var br = 0;
while ((br = stream.Read(buff, 0, buff.Length)) > 0)
{
if (br < buff.Length) // not a full sample, mute the rest
{
for (var i = br; i < buff.Length; i++)
{
buff[i] = 0;
}
}
await vnc.SendAsync(buff, 20, info.BitsPerSample);
}
}
await vnc.SendSpeakingAsync(false); // we're not speaking anymore
}
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 byte[] GenerateToByteArray(string text)
{
byte[] result;
using (MemoryStream streamDestination = new MemoryStream())
{
var speechSynthesizer = new SpeechSynthesizer();
speechSynthesizer.SetOutputToAudioStream(streamDestination, _formatInfo);
speechSynthesizer.Speak(text);
result = streamDestination.ToArray();
}
return(result);
}
private void SendTextToSpeech(object obj)
{
Connection c = (Connection)obj;
ConnectionStream stream = new ConnectionStream(c);
SpeechSynthesizer s = new SpeechSynthesizer();
using (MemoryStream memStream = new MemoryStream()) {
s.SetOutputToAudioStream(memStream, new SpeechAudioFormatInfo(EncodingFormat.ALaw,
8000, 8, 1, 8000, 1, null));
s.Speak(_textToSpeak);
memStream.Seek(0, SeekOrigin.Begin);
using (ConnectionWriter writer = new ConnectionWriter(stream)) {
writer.Reverse = true;
writer.Write(memStream);
}
}
}
/// <summary>
///Starts speech connector and starts speaking the prompt message asynchronously.
/// </summary>
private void StartSpeakAsync()
{
//start Speech synthesizer
if (AudioVideoCall.Flow.SpeechSynthesisConnector != null)
{
AudioVideoCall.Flow.SpeechSynthesisConnector.DetachFlow();
}
m_speechSynthesisConnector.AttachFlow(AudioVideoCall.Flow);
m_speechSynthesizer.SetOutputToAudioStream(m_speechSynthesisConnector, m_audioformat);
m_speechSynthesizer.SpeakCompleted += new EventHandler <SpeakCompletedEventArgs>(SpeechSynthesizer_SpeakCompleted);
m_speechSynthesisConnector.Start();
//If prompt is not set then speak Ssml pause.
if (m_pbMainPrompt == null)
{
m_pbMainPrompt = new PromptBuilder();
m_pbMainPrompt.AppendBreak();
}
m_speechSynthesizer.SpeakAsync(m_pbMainPrompt);
}
ReadAudio(string text, Message e)
{
var voiceChannel = e.Server.VoiceChannels.FirstOrDefault();
if (e.User.VoiceChannel != null && e.User.VoiceChannel.Name.ToLower() != "afk")
{
voiceChannel = e.User.VoiceChannel;
}
var _vClient = await voiceChannel.JoinAudio();
var channelCount = 2;
var OutFormat = new WaveFormat(48000, 16, channelCount);
SpeechSynthesizer synth = new SpeechSynthesizer();
MemoryStream ms = new MemoryStream();
synth.SetOutputToAudioStream(ms, new SpeechAudioFormatInfo(48000, AudioBitsPerSample.Sixteen, AudioChannel.Stereo));
synth.Speak(text);
byte[] bytes = ms.ToArray();
var test = new RawSourceWaveStream(ms, OutFormat);
test.Seek(0, SeekOrigin.Begin);
using (var resampler = new MediaFoundationResampler(test, OutFormat))
{
resampler.ResamplerQuality = 60;
int blockSize = OutFormat.AverageBytesPerSecond / 50;
byte[] buffer = new byte[blockSize];
int byteCount;
while ((byteCount = resampler.Read(buffer, 0, blockSize)) > 0)
{
if (byteCount < blockSize)
{
// Incomplete Frame
for (int i = byteCount; i < blockSize; i++)
{
buffer[i] = 0;
}
}
_vClient.Send(buffer, 0, blockSize);
}
}
}
public static void Speak(string text, string voice, int vol, int speed)
{
lock (speaklock) {
if (voiceStream == null)
{
voiceStream = audioClient.CreatePCMStream(AudioApplication.Voice, 128 * 1024);
}
SpeechSynthesizer tts = new SpeechSynthesizer();
tts.SelectVoice(voice);
tts.Volume = vol * 5;
tts.Rate = speed - 10;
MemoryStream ms = new MemoryStream();
tts.SetOutputToAudioStream(ms, formatInfo);
tts.Speak(text);
ms.Seek(0, SeekOrigin.Begin);
ms.CopyTo(voiceStream);
voiceStream.Flush();
}
}
public void StartWithWelcome(AudioVideoFlow flow, LyncServer server)
{
_flow = flow;
_server = server;
//attach speech synthasis to audio flow
_speechSynthesisConnector.AttachFlow(_flow);
_speechSynthesizer = new SpeechSynthesizer();
_speechSynthesizer.SetOutputToAudioStream(_speechSynthesisConnector, audioformat);
_speechSynthesizer.SelectVoice("Microsoft Hazel Desktop"); //slightly more english
var toneController = new ToneController(); //this is for the DTMF tones
toneController.AttachFlow(_flow);
_speechSynthesisConnector.Start();
_speechSynthesizer.Speak("Welcome to the UCMA IVR Demo!");
SpeakMenuOptions();
toneController.ToneReceived += toneController_ToneReceived;
}
public C2Voice(int zone)
{
myVoice = new SpeechSynthesizer();
rand = new Random();
InitializeVocabulary();
myZone = zone;
String host = "192.168.113." + (100 + myZone);
myRtpServer = new RTPServer(host, 1234);
myRtpServer.StartServer();
//ConnectSocket(host, 1234);
//myNetStream = new NetworkStream(myZoneSocket, true);
// SpeechAudioFormatInfo audioFormat = new SpeechAudioFormatInfo(24000, AudioBitsPerSample.Sixteen, AudioChannel.Mono);
// audioFormat.EncodingFormat = EncodingFormat.ALaw;
myVoice.SetOutputToAudioStream(myRtpServer.AudioStream, new SpeechAudioFormatInfo(24000, AudioBitsPerSample.Sixteen, AudioChannel.Mono));
// myVoice.SetOutputToAudioStream(myRtpServer.AudioStream, new SpeechAudioFormatType
// SpeechAudioFormatType;
myVoice.SetOutputToWaveStream(myRtpServer.AudioStream);
}
private void SendTextToSpeech(object obj) {
Connection c = (Connection)obj;
ConnectionStream stream = new ConnectionStream(c);
SpeechSynthesizer s = new SpeechSynthesizer();
using (MemoryStream memStream = new MemoryStream()) {
s.SetOutputToAudioStream(memStream, new SpeechAudioFormatInfo(EncodingFormat.ALaw,
8000, 8, 1, 8000, 1, null));
s.Speak(_textToSpeak);
memStream.Seek(0, SeekOrigin.Begin);
using (ConnectionWriter writer = new ConnectionWriter(stream)) {
writer.Reverse = true;
writer.Write(memStream);
}
}
}
private static byte[] GenerateVoiceAnnouncement(string announcement)
{
var synthesizer = new SpeechSynthesizer();
var waveStream = new MemoryStream();
var firstOrDefault = synthesizer.GetInstalledVoices()
.FirstOrDefault(x => x.VoiceInfo.Name.ToUpper().Contains("DAVID"));
if (firstOrDefault != null)
synthesizer.SelectVoice(
firstOrDefault
.VoiceInfo.Name);
synthesizer.SetOutputToAudioStream(waveStream,
new SpeechAudioFormatInfo(EncodingFormat.Pcm,
44100, 16, 2, 176400, 2, null));
synthesizer.Volume = 100;
synthesizer.Rate = 1;
synthesizer.Speak(announcement);
synthesizer.SetOutputToNull();
return waveStream.ToArray();
}
protected override void OnLoad(EventArgs e)
{
Visible = false;
ShowInTaskbar = false;
base.OnLoad(e);
/*
* Get all installed voices
*
*/
var voices = speech.GetInstalledVoices();
string voice = "";
foreach (InstalledVoice v in voices)
{
if (v.Enabled)
//voice = v.VoiceInfo.Name;
Console.WriteLine(v.VoiceInfo.Name);
}
queuetimer = new System.Timers.Timer(250);
queuetimer.Elapsed += (object sender, ElapsedEventArgs ev) =>
{
TTSRequest r;
if (Queue.TryDequeue(out r))
{
Console.WriteLine("dequeing off of concurrent queue...");
if (r.Interrupt)
{
// stop current TTS
if (IsSpeaking)
{
//speech.StopSpeaking();
}
if (IsSounding)
{
//sound.Stop();
if(sound.PlaybackState == PlaybackState.Playing) {
sound.Stop();
}
}
// clear queue
SpeechQueue.Clear();
}
if(!r.Reset) {
SpeechQueue.Enqueue(r);
}
RequestCount++;
}
var eventdata = new Hashtable();
eventdata.Add("ProcessedRequests", RequestCount);
eventdata.Add("QueuedRequests", SpeechQueue.Count);
eventdata.Add("IsSpeaking", IsSounding);
InstrumentationEvent blam = new InstrumentationEvent();
blam.EventName = "status";
blam.Data = eventdata;
NotifyGui(blam.EventMessage());
};
// when this timer fires, it will pull off of the speech queue and speak it
// the long delay also adds a little pause between tts requests.
speechtimer = new System.Timers.Timer(250);
speechtimer.Elapsed += (object sender, ElapsedEventArgs ev) =>
{
if (IsSpeaking.Equals(false))
{
if (SpeechQueue.Count > 0)
{
TTSRequest r = SpeechQueue.Dequeue();
Console.WriteLine("dequeuing off of speech queue");
IsSpeaking = true;
speechtimer.Enabled = false;
//speech.SpeakAsync(r.Text);
//using (speech = new SpeechSynthesizer()) {
speech = new SpeechSynthesizer();
speech.SpeakCompleted += speech_SpeakCompleted;
format = new SpeechAudioFormatInfo(EncodingFormat.ALaw, 8000, 8, 1, 1, 2, null);
//format = new SpeechAudioFormatInfo(11025, AudioBitsPerSample.Sixteen, AudioChannel.Mono);
// var si = speech.GetType().GetMethod("SetOutputStream", BindingFlags.Instance | BindingFlags.NonPublic);
stream = new MemoryStream();
//si.Invoke(speech, new object[] { stream, format, true, true });
//speech.SetOutputToWaveStream(stream);
speech.SetOutputToAudioStream(stream, format);
speech.SelectVoice(config.getVoice (r.Language, r.Voice));
int rate = (r.Speed * 2 - 10);
Console.WriteLine(rate);
try
{
speech.Rate = rate;
}
catch (ArgumentOutOfRangeException ex)
{
speech.Rate = 0;
}
speech.SpeakAsync(r.Text);
//}
synthesis.WaitOne();
speech.SpeakCompleted -= speech_SpeakCompleted;
speech.SetOutputToNull();
speech.Dispose();
//IsSpeaking = false;
IsSounding = true;
stream.Position = 0;
//WaveFormat.CreateCustomFormat(WaveFormatEncoding.WmaVoice9, 11025, 1, 16000, 2, 16)
using(RawSourceWaveStream reader = new RawSourceWaveStream(stream, WaveFormat.CreateALawFormat(8000, 1))) {
WaveStream ws = WaveFormatConversionStream.CreatePcmStream(reader);
//var waveProvider = new MultiplexingWaveProvider(new IWaveProvider[] { ws }, 4);
//waveProvider.ConnectInputToOutput(0, 3);
sound = new WaveOutEvent();
// set output device *before* init
Console.WriteLine("Output Device: " + OutputDeviceId);
sound.DeviceNumber = OutputDeviceId;
sound.Init(ws);
//sound.Init(waveProvider);
sound.PlaybackStopped += output_PlaybackStopped;
// Console.WriteLine("playing here " + ws.Length);
sound.Play();
}
playback.WaitOne();
//IsSounding = false;
speechtimer.Enabled = true;
}
}
};
queuetimer.Enabled = true;
queuetimer.Start();
speechtimer.Enabled = true;
speechtimer.Start();
InitHTTPServer();
}
