protected override void OnActivityResult(int requestCode, Result resultCode, Intent data)
{
base.OnActivityResult(requestCode, resultCode, data);
if (requestCode == SpeechRecognitionAndroidService.SpeakRequestCode)
{
if (resultCode == Result.Ok)
{
var matches = data.GetStringArrayListExtra(RecognizerIntent.ExtraResults);
if (matches.Count != 0)
{
Console.WriteLine(matches[0]);
SpeechRecognized?.Invoke(
this, new SpeechRecognizedEvent(recognizedText: matches[0]));
}
else
{
ReturnEmptyResponse();
}
}
else
{
ReturnEmptyResponse();
}
}
}
private void InvokeSpeechRecognized(String recognizedText)
{
if (SpeechRecognized != null)
{
SpeechRecognized.Invoke(this, recognizedText);
}
}
protected void RecEngine_SpeechRecognized(object sender, SpeechRecognizedEventArgs e)
{
VoiceCommand voiceCommand = _listCommands.FirstOrDefault(x => x.Commands.Contains(e.Result.Text))?.Copy();
if (voiceCommand == null)
{
return;
}
voiceCommand.DoSayResponse = false;
if (e.Result.Confidence < ConfidenceThreshold)
{
voiceCommand.ActionToPerform = VoiceCommandAction.DidntGetThat;
voiceCommand.Response = "I didn't get that";
}
switch (voiceCommand.ActionToPerform)
{
case VoiceCommandAction.GiveFeedback:
if (IsListening)
{
_isGivingFeedback = true;
SayResponseAsync(voiceCommand.Response);
}
break;
case VoiceCommandAction.StopGivingFeedback:
if (IsListening)
{
SayResponseAsync(voiceCommand.Response);
_isGivingFeedback = false;
}
break;
case VoiceCommandAction.DidntGetThat:
if (IsListening)
{
SayResponseAsync(voiceCommand.Response);
}
break;
case VoiceCommandAction.StartListening:
IsListening = true;
SayResponseAsync(voiceCommand.Response);
break;
case VoiceCommandAction.StopListening:
IsListening = false;
SayResponseAsync(voiceCommand.Response);
break;
default:
voiceCommand.DoSayResponse = true;
break;
}
if (!IsListening && voiceCommand.ActionToPerform != VoiceCommandAction.StopListening)
{
return;
}
SpeechRecognized?.Invoke(this, new ODSpeechRecognizedEventArgs(voiceCommand));
}
void _rg_SpeechRecognized(object sender, SpeechRecognizedEventArgs e)
{
if (e.Result == null)
{
return;
}
var ev = new RecognitionEventArgs(e.Result.Confidence, e.Result.Text);
SpeechRecognized?.Invoke(sender, ev);
}
private void HandleSpeechCompleted(object sender, RecognizeCompletedEventArgs eventArgs)
{
if (eventArgs.Result is null)
{
SpeechNotRecognized?.Invoke(this, new EventArgs());
}
else
{
var answer = new AnswerSelectedEventArgs
{
Result = true,
FieldName = currentAnswer.FieldName,
SelectedAnswer = GetProperResult(eventArgs),
};
SpeechRecognized?.Invoke(this, answer);
}
}
// Method called from the recognizer when a recognition has occurred.
// Only called for SpeechRecognition events, not SpeechRecognitionRejected.
internal void OnRecognitionInternal(SpeechRecognizedEventArgs eventArgs)
{
Debug.Assert(eventArgs.Result.Grammar == this);
SpeechRecognized?.Invoke(this, eventArgs);
}
private void ReturnEmptyResponse()
{
SpeechRecognized?.Invoke(this, new SpeechRecognizedEvent(recognizedText: ""));
}
public async Task TranslateAsync(YouTubeVideo youTubeVideo, string fromLanguage, IEnumerable <string> toLanguages)
{
// Declare the necessary directories and files variables
var outputPath = Path.Combine("Output", Guid.NewGuid().ToString());
var downloadFilePath = Path.Combine(outputPath, "input.mp4");
// StringBuilders for data to be passed to event subscriber
var tsb = new StringBuilder();
var osb = new StringBuilder();
var info = new StringBuilder();
var config = SpeechTranslationConfig.FromSubscription(
configuration["AzureSpeechTranslation:SubscriptionKey"], configuration["AzureSpeechTranslation:Region"]);
config.SpeechRecognitionLanguage = fromLanguage;
foreach (var language in toLanguages)
{
config.AddTargetLanguage(language);
}
var vidBytes = await youTubeVideo.GetBytesAsync();
// Before saving the video, create the directory
CreateOutputDirectory(outputPath);
// Save the video
await File.WriteAllBytesAsync(downloadFilePath, vidBytes);
// Extract the audio from the video to work on it
var wavAudioFile = await ExtractingWavAudioAsync(downloadFilePath);
var stopTranslation = new TaskCompletionSource <int>();
var lineCount = 1;
using (var audioInput = AudioConfig.FromWavFileInput(wavAudioFile))
{
using (var recognizer = new TranslationRecognizer(config, audioInput))
{
// Subscribes to events.
recognizer.Recognized += (s, e) =>
{
if (e.Result.Reason == ResultReason.TranslatedSpeech)
{
foreach (var element in e.Result.Translations)
{
var fromTime = TimeSpan.FromTicks(e.Result.OffsetInTicks);
var toTime = fromTime.Add(e.Result.Duration);
osb.AppendLine($"{lineCount}");
osb.AppendLine($"{fromTime.ToString(@"hh\:mm\:ss\.fff")} --> {toTime.ToString(@"hh\:mm\:ss\.fff")}");
osb.AppendLine(e.Result.Text);
osb.AppendLine();
tsb.AppendLine($"{lineCount}");
tsb.AppendLine($"{fromTime.ToString(@"hh\:mm\:ss\.fff")} --> {toTime.ToString(@"hh\:mm\:ss\.fff")}");
tsb.AppendLine(element.Value);
tsb.AppendLine();
var speechServicesEventArgs = SetSpeechServicesInformationArgs(fromLanguage, element.Key,
osb.ToString(), tsb.ToString());
osb.Clear();
tsb.Clear();
SpeechRecognized?.Invoke(this, speechServicesEventArgs);
}
lineCount++;
}
else if (e.Result.Reason == ResultReason.RecognizedSpeech)
{
info.AppendLine($"RECOGNIZED: Text={e.Result.Text}");
info.AppendLine($" Speech not translated.");
var speechServicesEventArgs = SetSpeechServicesInformationArgs(fromLanguage, information: info.ToString());
info.Clear();
SpeechRecognized?.Invoke(this, speechServicesEventArgs);
}
else if (e.Result.Reason == ResultReason.NoMatch)
{
info.AppendLine($"NOMATCH: Speech could not be recognized.");
var speechServicesEventArgs = SetSpeechServicesInformationArgs(fromLanguage, information: info.ToString());
info.Clear();
SpeechRecognized?.Invoke(this, speechServicesEventArgs);
}
};
recognizer.Canceled += (s, e) =>
{
info.AppendLine($"CANCELED: Reason={e.Reason}");
if (e.Reason == CancellationReason.Error)
{
info.AppendLine($"CANCELED: ErrorCode={e.ErrorCode}");
info.AppendLine($"CANCELED: ErrorDetails={e.ErrorDetails}");
info.AppendLine($"CANCELED: Did you update the subscription info?");
}
var speechServicesEventArgs = SetSpeechServicesInformationArgs(fromLanguage, information: info.ToString());
info.Clear();
SpeechCanceled?.Invoke(this, speechServicesEventArgs);
stopTranslation.TrySetResult(0);
};
recognizer.SpeechStartDetected += (s, e) =>
{
info.AppendLine("Speech start detected event.");
var speechServicesEventArgs = SetSpeechServicesInformationArgs(fromLanguage, information: info.ToString());
info.Clear();
SpeechStartDetected?.Invoke(this, speechServicesEventArgs);
};
recognizer.SpeechEndDetected += (s, e) =>
{
info.AppendLine("Speech end detected event.");
var speechServicesEventArgs = SetSpeechServicesInformationArgs(fromLanguage, information: info.ToString());
info.Clear();
SpeechEndDetected?.Invoke(this, speechServicesEventArgs);
};
recognizer.SessionStarted += (s, e) =>
{
info.AppendLine("Start translation...");
info.AppendLine("Session started event.");
var speechServicesEventArgs = SetSpeechServicesInformationArgs(fromLanguage, information: info.ToString());
info.Clear();
SpeechSessionStarted?.Invoke(this, speechServicesEventArgs);
};
recognizer.SessionStopped += (s, e) =>
{
info.AppendLine("Session stopped event.");
info.AppendLine("Stop translation.");
var speechServicesEventArgs = SetSpeechServicesInformationArgs(fromLanguage, information: info.ToString());
info.Clear();
SpeechSessionStopped?.Invoke(this, speechServicesEventArgs);
stopTranslation.TrySetResult(0);
};
// Starts continuous recognition. Uses StopContinuousRecognitionAsync() to stop recognition.
await recognizer.StartContinuousRecognitionAsync().ConfigureAwait(false);
// Waits for completion.
// Use Task.WaitAny to keep the task rooted.
Task.WaitAny(new[] { stopTranslation.Task });
// Stops translation.
await recognizer.StopContinuousRecognitionAsync().ConfigureAwait(false);
}
}
//Housekeeping
Directory.Delete(outputPath, true);
}
protected virtual void FireSpeechRecognized(string sentence)
{
SpeechRecognized?.Invoke(this, new SpeechRecognitionEventArgs(sentence));
}
protected void OnSpeechRecognized(SpeechRecognizedEvent e)
{
SpeechRecognized?.Invoke(this, e);
}
private void recognizer_SpeechRecognized(object sender, SpeechRecognizedEventArgs e)
{
Console.WriteLine("SpeechRecognized:" + e.Result.Text);
OnRecognized?.Invoke(e.Result.Text);
}
private void SpeechRecognizedProxy(object sender, SpeechRecognizedEventArgs e)
{
SpeechRecognized?.Invoke(this, e);
}
