/// <summary>
/// Configures NAudio for audio capture from either the default or specified audio device index
/// </summary>
/// <param name="deviceNumber">Desired Capture Device Index; Default is 0</param>
private void Initialize_NAudio(int deviceNumber = 0)
{
// If EventBasedAudio is already initialized, dispose
if (EventBasedAudio != null)
{
EventBasedAudio.Dispose();
}
EventBasedAudio = new WaveInEvent();
EventBasedAudio.DeviceNumber = deviceNumber;
EventBasedAudio.WaveFormat = WaveFormat.CreateCustomFormat(WaveFormatEncoding.Pcm, 8000, 1, 32000, 2, 16);
BufferedAudioProvider = new BufferedWaveProvider(EventBasedAudio.WaveFormat);
BufferedAudioProvider.BufferDuration = TimeSpan.FromSeconds(1);
Azure_NAudio_Callback = new NAudioCompatibleAudioCallback(ref BufferedAudioProvider, ref BufferedAudioProvider_LockObject);
EventBasedAudio.DataAvailable += (s, e) =>
{
lock (BufferedAudioProvider_LockObject)
{
BufferedAudioProvider.AddSamples(e.Buffer, 0, e.BytesRecorded);
}
if (AudioFileWriter != null)
{
AudioFileWriter.Write(e.Buffer, 0, e.BytesRecorded);
}
};
}
/// <summary>
/// Configures and Instantiates the Azure Speech-To-Text API; <seealso cref="Initialize_NAudio"/> must be called before using this function
/// </summary>
/// <param name="subscriptionKey">Valid Azure Subscription Key</param>
/// <param name="subscriptionRegion">Valid Azure Subscription Region</param>
/// <param name="CallbackFunction">Audio Source Callback Function</param>
private void Initialize_AzureSpeech(string subscriptionKey, string subscriptionRegion, NAudioCompatibleAudioCallback CallbackFunction)
{
// If SpeechRecognizer is already initialized, dispose
if (AzureSpeechRecognizer != null)
{
AzureSpeechRecognizer.Dispose();
}
// Configure SpeechConfig
SpeechConfig AzureSpeechConfig = SpeechConfig.FromSubscription(subscriptionKey, subscriptionRegion);
AzureSpeechConfig.SpeechRecognitionLanguage = "en-US";
AzureSpeechConfig.EnableDictation();
AzureSpeechConfig.EnableAudioLogging();
AzureSpeechConfig.SetProperty(PropertyId.Speech_LogFilename, @"C:\Users\Joe\source\repos\Monotone\Monotone\bin\x86\Debug x86\log.txt");
AzureSpeechConfig.SetProfanity(ProfanityOption.Raw);
AzureSpeechConfig.OutputFormat = OutputFormat.Detailed;
// Configure AudioConfig
AudioConfig AzureAudioConfig = AudioConfig.FromStreamInput(CallbackFunction,
AudioStreamFormat.GetWaveFormatPCM((uint)EventBasedAudio.WaveFormat.SampleRate,
(byte)EventBasedAudio.WaveFormat.BitsPerSample,
(byte)EventBasedAudio.WaveFormat.Channels));
// Initialize SpeechRecognizer
AzureSpeechRecognizer = new SpeechRecognizer(AzureSpeechConfig, AzureAudioConfig);
// Configure Recognizer Events
Azure_StopRecognition = new TaskCompletionSource <int>();
//AzureSpeechRecognizer.Recognizing += (s, e) =>
//{
// Debug.WriteLine($"RECOGNIZING: Text ={e.Result.Text}");
//};
AzureSpeechRecognizer.Recognized += (s, e) =>
{
if (e.Result.Reason == ResultReason.RecognizedSpeech)
{
monotone_RecordTranscribe1.AddEntryLine(e.Result.Text, (timeIndex_Stopwatch.Elapsed - e.Result.Duration));
}
};
AzureSpeechRecognizer.Canceled += (s, e) =>
{
if (e.Reason == CancellationReason.Error)
{
MessageBox.Show($"Canceled due to Error {e.ErrorCode}\n\n" + $"Details: {e.ErrorDetails}\n\n" + $"Did you update your subscription info?",
$"Azure Speech Recognition Error {e.ErrorCode}", MessageBoxButtons.OK, MessageBoxIcon.Error);
}
StopRecording(this, new EventArgs());
};
AzureSpeechRecognizer.SessionStopped += (s, e) =>
{
Debug.WriteLine("Session Stopped");
StopRecording(this, new EventArgs());
};
AzureSpeechRecognizer.SessionStarted += (s, e) =>
{
Debug.WriteLine("Session Start");
};
AzureSpeechRecognizer.SessionStopped += (s, e) =>
{
Debug.WriteLine("Session Stopped");
};
}
