public async Task StartSpeechRecognitionAsync()
{
SpeechTranslationConfig config = GetRecognizerConfig();
if (config == null)
{
return;
}
ResetState();
DisposeRecognizer();
DeviceInformation microphoneInput = await Util.GetDeviceInformation(DeviceClass.AudioCapture, SettingsHelper.Instance.MicrophoneName);
using (AudioConfig audioConfig = AudioConfig.FromMicrophoneInput(microphoneInput.Id))
{
translationRecognizer = audioConfig != null ? new TranslationRecognizer(config, audioConfig) : new TranslationRecognizer(config);
translationRecognizer.Recognizing += OnTranslateRecognizing;
translationRecognizer.Recognized += OnTranslateRecognized;
translationRecognizer.Canceled += OnTranslateCanceled;
translationRecognizer.SessionStarted += (s, e) =>
{
recognizeCancellationTokenSource = new CancellationTokenSource();
};
await translationRecognizer.StartContinuousRecognitionAsync();
}
}
private void SetupTranscriptionAndTranslationService()
{
try
{
var lCognitiveKey = _settings.AzureCognitiveKey;
var lCognitiveRegion = _settings.AzureCognitiveRegion;
_eventPublisher.Publish("MySTT Setup", $"Got region: {lCognitiveRegion}, key starting from: {lCognitiveKey??lCognitiveKey.Substring(0, lCognitiveKey.Length /2)}");
this.mTransSpeechConfig = SpeechTranslationConfig.FromSubscription(lCognitiveKey, lCognitiveRegion);
var fromLanguage = "en-US";
var toLanguages = new List <string> {
"el-GR"
};
//var toLanguages = new List<string> { "ru-RU" };
this.mTransSpeechConfig.SpeechRecognitionLanguage = fromLanguage;
toLanguages.ForEach(this.mTransSpeechConfig.AddTargetLanguage);
this.mInputStream = AudioInputStream.CreatePushStream(AudioStreamFormat.GetWaveFormatPCM(SAMPLESPERSECOND, BITSPERSAMPLE, NUMBEROFCHANNELS));
this.mAudioConfig = AudioConfig.FromStreamInput(this.mInputStream);
this.mTranslationRecognizer = new TranslationRecognizer(this.mTransSpeechConfig, this.mAudioConfig);
this.mTranslationRecognizer.Recognizing += this.MSpeechRecognizer_Recognizing;
this.mTranslationRecognizer.Recognized += this.MSpeechRecognizer_Recognized;
this.mTranslationRecognizer.SpeechEndDetected += this.MSpeechRecognizer_SpeechEndDetected;
this.StartRecognisionIfNeeded();
}
catch (Exception ex)
{
_eventPublisher.Publish("MySTT Setup - Failed", $"Failed to initialize: {ex.Message}");
}
}
public AzureCognitiveService(IDictionary <AzureServiceType, AzureServiceAuthorization> servicesAuthorization, ISDKLogger logger)
{
_logger = logger;
_servicesAuthorization = servicesAuthorization;
foreach (KeyValuePair <AzureServiceType, AzureServiceAuthorization> auth in _servicesAuthorization)
{
if (auth.Key == AzureServiceType.ComputerVision)
{
_computerVisionClient = new ComputerVisionClient(
new ApiKeyServiceClientCredentials(auth.Value.SubscriptionKey),
new System.Net.Http.DelegatingHandler[] { });
_computerVisionClient.Endpoint = auth.Value.Endpoint;
_availableServices.Add(AzureServiceType.ComputerVision);
}
else if (auth.Key == AzureServiceType.Speech)
{
_speechConfig = SpeechConfig.FromSubscription(auth.Value.SubscriptionKey, auth.Value.Region);
_speechTranslationConfig = SpeechTranslationConfig.FromSubscription(auth.Value.SubscriptionKey, auth.Value.Region);
SetProfanityOption(AzureProfanitySetting);
_speechTranslationConfig.SpeechSynthesisVoiceName = CurrentSpeakingVoice;
_availableServices.Add(AzureServiceType.Speech);
}
}
}
private SpeechTranslationConfig GetRecognizerConfig()
{
var translationLanguageCodes = new List <string>();
if (this.firstTranslateLanguageCombobox.SelectedValue is SupportedLanguage firstLanguage)
{
translationLanguageCodes.Add(firstLanguage.Code);
}
if (this.secondTranslateLanguageCombobox.SelectedValue is SupportedLanguage secondLanguage)
{
translationLanguageCodes.Add(secondLanguage.Code);
}
if (!(this.inputLanguageCombobox.SelectedValue is SupportedLanguage language) || !translationLanguageCodes.Any())
{
return(null);
}
var speechTranslationConfig = SpeechTranslationConfig.FromEndpoint(GetSpeechTranslationEndpoint(SettingsHelper.Instance.SpeechApiEndpoint), SettingsHelper.Instance.SpeechApiKey);
speechTranslationConfig.SpeechRecognitionLanguage = language.Code;
foreach (string code in translationLanguageCodes)
{
speechTranslationConfig.AddTargetLanguage(code);
}
return(speechTranslationConfig);
}
private SpeechTranslationConfig GetSpeechClient()
{
var subscriptionKey = configurationReader.GetKey();
var region = configurationReader.GetRegion();
return(SpeechTranslationConfig.FromSubscription(subscriptionKey, region));
}
public static async Task <String> TranslationContinuousRecognitionAsync()
{
var config = SpeechTranslationConfig.FromSubscription("a8c500c9ef9f421e977486a17f0adcca", "westeurope");
string fromLanguage = "pl-PL";
config.SpeechRecognitionLanguage = fromLanguage;
config.AddTargetLanguage("pl");
string result = "";
// Sets voice name of synthesis output.
const string PolishVoice = "pl-PL";
config.VoiceName = PolishVoice;
// Creates a translation recognizer using microphone as audio input.
using (var recognizer = new TranslationRecognizer(config))
{
// Subscribes to events.
recognizer.Recognized += async(s, e) =>
{
if (e.Result.Reason == ResultReason.TranslatedSpeech)
{
await recognizer.StopContinuousRecognitionAsync();
result = e.Result.Text;
}
};
return(result);
}
}
private void OnDestroy()
{
if (this.config != null)
{
this.recognizer.Dispose();
this.config = null;
}
}
public void Initialize(string subscriptionKey, string region)
{
subscriptionKey.EnsureIsNotNull(nameof(subscriptionKey));
subscriptionKey.EnsureIsNotNull(nameof(region));
speechConfiguration = SpeechTranslationConfig.FromSubscription(subscriptionKey, region);
speechConfiguration.OutputFormat = OutputFormat.Detailed;
SendMessage($"Created the SpeechConfiguration with {subscriptionKey} | {region}");
}
public static async Task TranslateSpeechToText()
{
// Creates an instance of a speech translation config with specified subscription key and service region.
// Replace with your own subscription key and service region (e.g., "westus").
var config = SpeechTranslationConfig.FromSubscription("311b76d8841344b6a277dbd8401611dc", "westus");
// Sets source and target languages.
// Replace with the languages of your choice, from list found here: https://aka.ms/speech/sttt-languages
string fromLanguage = "en-US";
config.SpeechRecognitionLanguage = fromLanguage;
config.AddTargetLanguage("de");
config.AddTargetLanguage("fr");
// Creates a translation recognizer using the default microphone audio input device.
using (var recognizer = new TranslationRecognizer(config))
{
// Starts translation, and returns after a single utterance is recognized. The end of a
// single utterance is determined by listening for silence at the end or until a maximum of 15
// seconds of audio is processed. The task returns the recognized text as well as the translation.
// Note: Since RecognizeOnceAsync() returns only a single utterance, it is suitable only for single
// shot recognition like command or query.
// For long-running multi-utterance recognition, use StartContinuousRecognitionAsync() instead.
Console.WriteLine("Say something...");
var result = await recognizer.RecognizeOnceAsync();
// Checks result.
if (result.Reason == ResultReason.TranslatedSpeech)
{
Console.WriteLine($"RECOGNIZED '{fromLanguage}': {result.Text}");
foreach (var element in result.Translations)
{
Console.WriteLine($"TRANSLATED into '{element.Key}': {element.Value}");
}
}
else if (result.Reason == ResultReason.RecognizedSpeech)
{
Console.WriteLine($"RECOGNIZED '{fromLanguage}': {result.Text} (text could not be translated)");
}
else if (result.Reason == ResultReason.NoMatch)
{
Console.WriteLine($"NOMATCH: Speech could not be recognized.");
}
else if (result.Reason == ResultReason.Canceled)
{
var cancellation = CancellationDetails.FromResult(result);
Console.WriteLine($"CANCELED: Reason={cancellation.Reason}");
if (cancellation.Reason == CancellationReason.Error)
{
Console.WriteLine($"CANCELED: ErrorCode={cancellation.ErrorCode}");
Console.WriteLine($"CANCELED: ErrorDetails={cancellation.ErrorDetails}");
Console.WriteLine($"CANCELED: Did you update the subscription info?");
}
}
}
}
public TranslationEngine(IConfiguration config, IHubContext <TranslationHub> hub)
{
_hub = hub;
_config = config;
_translationConfig = SpeechTranslationConfig.FromSubscription(_config["SUBSCRIPTION_KEY"], _config["REGION"]);
_speechConfig = SpeechTranslationConfig.FromSubscription(_config["SUBSCRIPTION_KEY"], _config["REGION"]);
_audioInput = AudioConfig.FromStreamInput(_inputStream);
_audioOutputStream = AudioOutputStream.CreatePullStream();
_output = AudioConfig.FromStreamOutput(_audioOutputStream);
}
static async Task Main(string[] args)
{
try
{
// Get config settings from AppSettings
IConfigurationBuilder builder = new ConfigurationBuilder().AddJsonFile("appsettings.json");
IConfigurationRoot configuration = builder.Build();
string cogSvcKey = configuration["CognitiveServiceKey"];
string cogSvcRegion = configuration["CognitiveServiceRegion"];
// Set a dictionary of supported voices
var voices = new Dictionary <string, string>
{
["fr"] = "fr-FR-Julie",
["es"] = "es-ES-Laura",
["hi"] = "hi-IN-Kalpana"
};
// Configure translation
translationConfig = SpeechTranslationConfig.FromSubscription(cogSvcKey, cogSvcRegion);
translationConfig.SpeechRecognitionLanguage = "en-US";
Console.WriteLine("Ready to translate from " + translationConfig.SpeechRecognitionLanguage);
string targetLanguage = "";
while (targetLanguage != "quit")
{
Console.WriteLine("\nEnter a target language\n fr = French\n es = Spanish\n hi = Hindi\n Enter anything else to stop\n");
targetLanguage = Console.ReadLine().ToLower();
// Check if the user has requested a language that this app supports
if (voices.ContainsKey(targetLanguage))
{
// Because the synthesised speech event only supports 1:1 translation, we'll remove any languages already in the translationconfig
if (translationConfig.TargetLanguages.Count > 1)
{
foreach (string language in translationConfig.TargetLanguages)
{
translationConfig.RemoveTargetLanguage(language);
}
}
// and add the requested one in
translationConfig.AddTargetLanguage(targetLanguage);
translationConfig.VoiceName = voices[targetLanguage];
await Translate(targetLanguage);
}
else
{
targetLanguage = "quit";
}
}
}
catch (Exception ex) { Console.WriteLine(ex.Message); }
}
static void Main()
{
var config = SpeechTranslationConfig.FromSubscription(ConfigurationManager.AppSettings.Get("TTSKey"), ConfigurationManager.AppSettings.Get("Region"));
//RecognizeOnceSpeechAsync(config).Wait();
//Translate.RecognizeLng().Wait();
Translate.TranslationContinuousRecognitionAsync(config).Wait();
//TTS(config).Wait();
Console.WriteLine("Please press a key to continue.");
Console.ReadLine();
}
public void Initialize(string subscriptionKey, string region, InputSourceType inputSource, string wavFilename)
{
subscriptionKey.EnsureIsNotNull(nameof(subscriptionKey));
subscriptionKey.EnsureIsNotNull(nameof(region));
speechConfiguration = SpeechTranslationConfig.FromSubscription(subscriptionKey, region);
speechConfiguration.OutputFormat = OutputFormat.Detailed;
SendMessage($"Created the SpeechConfiguration with {subscriptionKey} | {region}");
audioConfig = GetAudioConfig(inputSource, wavFilename);
}
public async Task <ActionResult> translateAsync()
{
string fromLanguage = "en-US";
const string GermanVoice = "Microsoft Server Speech Text to Speech Voice (de-DE, Hedda)";
var config = SpeechTranslationConfig.FromSubscription("ae9492aae8044a4c888a45a45e957d83", "westus");
config.SpeechRecognitionLanguage = fromLanguage;
config.VoiceName = GermanVoice;
config.AddTargetLanguage("de");
using (var recognizer = new TranslationRecognizer(config))
{
recognizer.Recognizing += (s, e) =>
{
foreach (var element in e.Result.Translations)
{
ViewBag.message = element.Value;
}
};
recognizer.Recognized += (s, e) =>
{
if (e.Result.Reason == ResultReason.TranslatedSpeech)
{
foreach (var element in e.Result.Translations)
{
ViewBag.message = element.Value;
}
}
else if (e.Result.Reason == ResultReason.RecognizedSpeech)
{
foreach (var element in e.Result.Translations)
{
ViewBag.message = element.Value;
}
}
else if (e.Result.Reason == ResultReason.NoMatch)
{
}
};
await recognizer.RecognizeOnceAsync();
return(View("Index"));
}
}
public static async Task Translate(string subscriptionKey, string region, string inputFilename, string fromLanguage, IEnumerable <string> targetLanguages, Voice voice, string outputFilename)
{
if (!outputFilename.EndsWith(".wav") && !outputFilename.EndsWith(".mp3"))
{
throw new ArgumentOutOfRangeException(paramName: nameof(outputFilename), message: "Output filename must have '.wav' or '.mp3' extension");
}
var config = SpeechTranslationConfig.FromSubscription(subscriptionKey, region);
var wavBytes = await GetWAVFromFile(inputFilename);
await config.TranslationWithFileAsync(wavBytes, fromLanguage, targetLanguages, voice, outputFilename);
}
/// <summary>
/// Initializes the config object with subscription key and region
/// Initializes the recognizer object with a TranslationRecognizer
/// Subscribes the recognizer to recognition Event Handlers
/// If recognition is running, starts a thread which stops the recognition
/// </summary>
private void CreateRecognizer()
{
this.config = SpeechTranslationConfig.FromSubscription(SubscriptionKey, Region);
this.config.SpeechRecognitionLanguage = FromLanguage;
this.config.VoiceName = voice;
ToLanguages.ForEach(l => this.config.AddTargetLanguage(l));
this.recognizer = new TranslationRecognizer(this.config);
this.recognizer.Recognizing += this.OnRecognizingEventHandler;
this.recognizer.Recognized += this.OnRecognizedEventHandler;
this.recognizer.Synthesizing += this.OnSynthesizingEventHandler;
this.recognizer.Canceled += this.OnCanceledEventHandler;
}
void Start()
{
if (outputText == null)
{
UnityEngine.Debug.LogError("outputText property is null! Assign a UI Text element to it.");
}
else if (recoButton == null)
{
_message = "recoButton property is null! Assign a UI Button to it.";
UnityEngine.Debug.LogError(_message);
}
else
{
// Continue with normal initialization, Text and Button objects are present.
#if PLATFORM_ANDROID
// Request to use the microphone, cf.
// https://docs.unity3d.com/Manual/android-RequestingPermissions.html
message = "Waiting for mic permission";
if (!Permission.HasUserAuthorizedPermission(Permission.Microphone))
{
Permission.RequestUserPermission(Permission.Microphone);
}
#elif PLATFORM_IOS
if (!Application.HasUserAuthorization(UserAuthorization.Microphone))
{
Application.RequestUserAuthorization(UserAuthorization.Microphone);
}
#else
_micPermissionGranted = true;
_message = "Click button to recognize speech";
#endif
_config = SpeechTranslationConfig.FromSubscription(SubscriptionKey, SubscriptionRegion);
_config.SpeechRecognitionLanguage = "es-US";
_config.AddTargetLanguage("en-US");
_pushStream = AudioInputStream.CreatePushStream();
_audioInput = AudioConfig.FromStreamInput(_pushStream);
_recognizer = new TranslationRecognizer(_config, _audioInput);
_recognizer.Recognizing += RecognizingHandler;
_recognizer.Recognized += RecognizedHandler;
_recognizer.Canceled += CanceledHandler;
foreach (var device in Microphone.devices)
{
Debug.Log("DeviceName: " + device);
}
_audioSource = GameObject.Find("AudioSource").GetComponent <AudioSource>();
}
}
private void CreateRecognizer()
{
//Microsoft.CognitiveServices.Speech.Internal.carbon_csharp a = new car();
this.config = SpeechTranslationConfig.FromSubscription(SubscriptionKey, Region);
this.config.SpeechRecognitionLanguage = FromLanguage;
this.config.VoiceName = voice;
ToLanguages.ForEach(l => this.config.AddTargetLanguage(l));
this.recognizer = new TranslationRecognizer(this.config);
this.recognizer.Recognizing += this.OnRecognizingEventHandler;
this.recognizer.Recognized += this.OnRecognizedEventHandler;
this.recognizer.Synthesizing += this.OnSynthesizingEventHandler;
this.recognizer.Canceled += this.OnCanceledEventHandler;
}
public async Task <(ResultReason, string)> ListenAsync()
{
var sourceLanguageConfigs = new SourceLanguageConfig[]
{
SourceLanguageConfig.FromLanguage("en-US"),
SourceLanguageConfig.FromLanguage("it-IT")
};
var config = SpeechTranslationConfig.FromSubscription(Config.Key, Config.Region);
var autoDetectSourceLanguageConfig = AutoDetectSourceLanguageConfig.FromSourceLanguageConfigs(sourceLanguageConfigs);
using var recognizer = new SpeechRecognizer(config, autoDetectSourceLanguageConfig);
var result = await recognizer.RecognizeOnceAsync();
return(result.Reason switch
{
ResultReason.RecognizedSpeech => (ResultReason.RecognizedSpeech, result.Text),
_ => (ResultReason.NoMatch, null)
});
private void Init(string from, string to)
{
this.toLanguage = to;
Profile = MediaEncodingProfile.CreateWav(AudioEncodingQuality.Low);
Profile.Audio = AudioEncodingProperties.CreatePcm(16000, 1, 16);
byte channels = 1;
byte bitsPerSample = 16;
uint samplesPerSecond = 16000; // or 8000
var audioFormat = AudioStreamFormat.GetWaveFormatPCM(samplesPerSecond, bitsPerSample, channels);
// Init Push Stream
pushStream = AudioInputStream.CreatePushStream(audioFormat);
if (from == to)
{
var config = SpeechConfig.FromSubscription(apiKey, region);
config.SpeechRecognitionLanguage = from;
speechRecognizer = new SpeechRecognizer(config, AudioConfig.FromStreamInput(pushStream));
speechRecognizer.Recognizing += RecognisingSpeechHandler;
speechRecognizer.Recognized += RecognisingSpeechHandler;
speechRecognizer.SessionStarted += (sender, args) => this.RecognisionStarted?.Invoke();
speechRecognizer.SessionStopped += (sender, args) => this.RecognisionStopped?.Invoke();
}
else
{
var config = SpeechTranslationConfig.FromSubscription(apiKey, region);
config.SpeechRecognitionLanguage = from;
config.AddTargetLanguage(to);
translationRecognizer = new TranslationRecognizer(config, AudioConfig.FromStreamInput(pushStream));
translationRecognizer.SessionStarted += (sender, args) => this.RecognisionStarted?.Invoke();
translationRecognizer.SessionStopped += (sender, args) => this.RecognisionStopped?.Invoke();
translationRecognizer.Recognizing += RecognisingTranslationHandler;
translationRecognizer.Recognized += RecognisingTranslationHandler;
}
}
public async void ButtonClick()
{
Debug.Log("Onclick fired");
var translationConfig = SpeechTranslationConfig.FromSubscription(SpeechServiceSubscriptionKey, SpeechServiceRegion);
translationConfig.SpeechRecognitionLanguage = "en-US";
translationConfig.AddTargetLanguage("fr");
using (var recognizer = new TranslationRecognizer(translationConfig))
{
Debug.Log("Creating recognizer");
lock (threadLocker)
{
waitingforReco = true;
}
var result = await recognizer.RecognizeOnceAsync().ConfigureAwait(false);
if (result.Reason == ResultReason.TranslatedSpeech)
{
recognizedString = result.Text;
Debug.Log("Text: " + recognizedString);
foreach (var element in result.Translations)
{
translatedString = element.Value;
}
}
else if (result.Reason == ResultReason.NoMatch)
{
recognizedString = "NOMATCH: Speech could not be recognized.";
}
else if (result.Reason == ResultReason.Canceled)
{
var cancellation = CancellationDetails.FromResult(result);
recognizedString = $"CANCELED: Reason={cancellation.Reason} ErrorDetails={cancellation.ErrorDetails}";
}
lock (threadLocker)
{
waitingforReco = false;
}
}
}
private async void StartRecognition()
{
if (isRecognitionStarted)
{
return;
}
Debug.Log("start recognition");
string fromLang;
string toLang;
if (modeList.value == 0)
{
fromLang = "ja-JP";
toLang = "en";
}
else
{
fromLang = "en-US";
toLang = "ja";
}
Debug.Log("mode : " + fromLang + " -> " + toLang);
var config = SpeechTranslationConfig.FromSubscription(apiKeyInputField.text, apiRegionInputField.text);
config.SpeechRecognitionLanguage = fromLang;
config.AddTargetLanguage(toLang);
recognizer = new TranslationRecognizer(config);
recognizer.Canceled += CanceledHandler;
recognizer.SessionStarted += SessionStartedHandler;
recognizer.SessionStopped += SessionStoppedHandler;
recognizer.SpeechStartDetected += SpeechStartDetectedHandler;
recognizer.SpeechEndDetected += SpeechEndDetectedHandler;
recognizer.Recognizing += RecognizingHandler;
recognizer.Recognized += RecognizedHandler;
await recognizer.StartContinuousRecognitionAsync().ConfigureAwait(false);
isRecognitionStarted = true;
isRecognitionStateChanged = true;
}
public async Task SpeechRecognitionFromFileAsync(StorageFile file)
{
SpeechTranslationConfig config = GetRecognizerConfig();
if (config == null)
{
return;
}
ResetState();
stopRecognitionTaskCompletionSource = new TaskCompletionSource <int>();
using (var audioInput = AudioConfig.FromWavFileInput(file.Path))
{
using (var recognizer = new TranslationRecognizer(config, audioInput))
{
recognizer.Recognizing += OnTranslateRecognizing;
recognizer.Recognized += OnTranslateRecognized;
recognizer.Canceled += OnTranslateCanceled;
recognizer.SessionStarted += (s, e) =>
{
recognizeCancellationTokenSource = new CancellationTokenSource();
};
recognizer.SessionStopped += (s, e) =>
{
if (recognizeCancellationTokenSource != null && recognizeCancellationTokenSource.Token.CanBeCanceled)
{
recognizeCancellationTokenSource.Cancel();
}
stopRecognitionTaskCompletionSource.TrySetResult(0);
};
// Starts continuous recognition.
await recognizer.StartContinuousRecognitionAsync().ConfigureAwait(false);
// Waits for completion.
await stopRecognitionTaskCompletionSource.Task.ConfigureAwait(false);
// Stops recognition.
await recognizer.StopContinuousRecognitionAsync().ConfigureAwait(false);
}
}
}
void CreateTranslationRecognizer()
{
if (translator == null)
{
SpeechTranslationConfig config = SpeechTranslationConfig.FromSubscription(lunarcomController.SpeechServiceAPIKey, lunarcomController.SpeechServiceRegion);
config.SpeechRecognitionLanguage = fromLanguage;
config.AddTargetLanguage(toLanguage);
translator = new TranslationRecognizer(config);
if (translator != null)
{
translator.Recognizing += HandleTranslatorRecognizing;
translator.Recognized += HandleTranslatorRecognized;
translator.Canceled += HandleTranslatorCanceled;
translator.SessionStarted += HandleTranslatorSessionStarted;
translator.SessionStopped += HandleTranslatorSessionStopped;
}
}
}
private SpeechTranslationConfig createSpeechTranslationConfig(String logId, Key key, string sourceLanguage, List <string> languages)
{
SpeechTranslationConfig speechConfig = SpeechTranslationConfig.FromSubscription(key.ApiKey, key.Region);
speechConfig.RequestWordLevelTimestamps();
if (!IsSupportedRecognition(sourceLanguage))
{
_logger.LogError($"{logId}: !!!! Unknown recognition language ({sourceLanguage})! Recogition may fail ...");
}
speechConfig.SpeechRecognitionLanguage = sourceLanguage;
_logger.LogInformation($"{logId}: Requested output languages: { String.Join(",", languages) }, source = ({sourceLanguage})");
String shortCodeSource = sourceLanguage.Split('-')[0].ToLower();
foreach (var language in languages)
{
String shortCodeTarget = language.Split('-')[0].ToLower();
if (shortCodeSource == shortCodeTarget)
{
continue;
}
if (IsSupportedTranslation(language))
{
_logger.LogInformation($"{logId}: Adding target {language}");
speechConfig.AddTargetLanguage(language);
}
else
{
_logger.LogWarning($"{logId}: Skipping unsupported target {language}");
}
}
speechConfig.OutputFormat = OutputFormat.Detailed;
return(speechConfig);
}
public async Task <MSTResult> RecognitionWithVideoStreamAsync(string logId, string videoFilePath, Key key, Dictionary <string, List <Caption> > captions, string sourceLanguage, Dictionary <string, TimeSpan> startAfterMap)
{
List <string> outputLanguages = startAfterMap.Keys.ToList <string>();
TimeSpan restartOffset = startAfterMap.Any() ? startAfterMap.Values.Min() : TimeSpan.Zero;
_logger.LogInformation($"{logId}:RecognitionWithVideoStreamAsync restartOffset=({restartOffset.TotalSeconds}) seconds");
var trimmedAudioFile = await createTrimmedAudioFileAsync(videoFilePath, (float)restartOffset.TotalSeconds);
try
{
SpeechTranslationConfig speechConfig = createSpeechTranslationConfig(logId, key, sourceLanguage, outputLanguages);
var result = await performRecognitionAsync(logId, trimmedAudioFile.FilePath, speechConfig, restartOffset,
sourceLanguage, captions, startAfterMap);
return(result);
}
finally
{
QuietDelete(trimmedAudioFile.FilePath);
}
}
/// <summary>
/// Creates a class-level Translation Recognizer for a specific language using Azure credentials
/// and hooks-up lifecycle & recognition events. Translation can be enabled with one or more target
/// languages translated simultaneously
/// </summary>
void CreateTranslationRecognizer()
{
Debug.Log("Creating Translation Recognizer.");
recognizedString = "Initializing speech recognition with translation, please wait...";
if (translator == null)
{
SpeechTranslationConfig config = SpeechTranslationConfig.FromSubscription(SpeechServiceAPIKey, SpeechServiceRegion);
config.SpeechRecognitionLanguage = fromLanguage;
if (Languages1.captionText.text.Length > 0)
{
config.AddTargetLanguage(ExtractLanguageCode(Languages1.captionText.text));
}
if (Languages2.captionText.text.Length > 0)
{
config.AddTargetLanguage(ExtractLanguageCode(Languages2.captionText.text));
}
if (Languages3.captionText.text.Length > 0)
{
config.AddTargetLanguage(ExtractLanguageCode(Languages3.captionText.text));
}
translator = new TranslationRecognizer(config);
if (translator != null)
{
translator.Recognizing += RecognizingTranslationHandler;
translator.Recognized += RecognizedTranslationHandler;
translator.SpeechStartDetected += SpeechStartDetectedHandler;
translator.SpeechEndDetected += SpeechEndDetectedHandler;
translator.Canceled += CanceledTranslationHandler;
translator.SessionStarted += SessionStartedHandler;
translator.SessionStopped += SessionStoppedHandler;
}
}
Debug.Log("CreateTranslationRecognizer exit");
}
static async Task CreateConversationAsync()
{
// Replace with your own subscription key and service region (e.g., "westus").
string subscriptionKey = "YourSubscriptionKey";
string region = "YourServiceRegion";
// Change this to match the language you are speaking. You can find the full list of supported
// speech language codes here: https://docs.microsoft.com/azure/cognitive-services/speech-service/language-support
string fromLanguage = "en-US";
// Change this to the language you would like to translate the transcriptions to. You can find
// the full list of supported translation languages here: https://aka.ms/speech/sttt-languages
string toLanguage = "de";
// Set this to the display name you want for the conversation host
string displayName = "The host";
// Creates an instance of a speech config with specified subscription key and service region.
var config = SpeechTranslationConfig.FromSubscription(subscriptionKey, region);
config.SpeechRecognitionLanguage = fromLanguage;
config.AddTargetLanguage(toLanguage);
// Create the conversation object you'll need to manage the conversation
using (var conversation = await Conversation.CreateConversationAsync(config).ConfigureAwait(false))
{
// Start the conversation so you and others can join
await conversation.StartConversationAsync().ConfigureAwait(false);
// Get the conversation ID. It will be up to your scenario to determine how this is shared
// with other participants.
string conversationId = conversation.ConversationId;
Console.WriteLine($"CONVERSATION: Created a new conversation with ID '{conversationId}'");
// At this point, you can use the conversation object to manage the conversation. For example,
// to mute everyone else in the room you can call this method:
await conversation.MuteAllParticipantsAsync().ConfigureAwait(false);
// Configure which audio source you want to use. In this case we will use your default microphone
var audioConfig = AudioConfig.FromDefaultMicrophoneInput();
// Create the conversation translator you'll need to send audio, send IMs, and receive conversation events
using (var conversationTranslator = new ConversationTranslator(audioConfig))
{
// You should connect all the event handlers you need at this point
conversationTranslator.SessionStarted += (s, e) =>
{
Console.WriteLine($"SESSION STARTED: {e.SessionId}");
};
conversationTranslator.SessionStopped += (s, e) =>
{
Console.WriteLine($"SESSION STOPPED: {e.SessionId}");
};
conversationTranslator.Canceled += (s, e) =>
{
Console.WriteLine($"CANCELED: Reason={e.Reason}");
switch (e.Reason)
{
case CancellationReason.EndOfStream:
Console.WriteLine($"CANCELED: End of audio reached");
break;
case CancellationReason.Error:
Console.WriteLine($"CANCELED: ErrorCode= {e.ErrorCode}");
Console.WriteLine($"CANCELED: ErrorDetails= {e.ErrorDetails}");
break;
}
};
conversationTranslator.ConversationExpiration += (s, e) =>
{
Console.WriteLine($"CONVERSATION: Will expire in {e.ExpirationTime.TotalMinutes} minutes");
};
conversationTranslator.ParticipantsChanged += (s, e) =>
{
Console.Write("PARTICIPANTS: The following participant(s) have ");
switch (e.Reason)
{
case ParticipantChangedReason.JoinedConversation:
Console.Write("joined");
break;
case ParticipantChangedReason.LeftConversation:
Console.Write("left");
break;
case ParticipantChangedReason.Updated:
Console.Write("been updated");
break;
}
Console.WriteLine(":");
foreach (var participant in e.Participants)
{
Console.WriteLine($"\tPARTICIPANT: {participant.DisplayName}");
}
};
conversationTranslator.TextMessageReceived += (s, e) =>
{
Console.WriteLine($"TEXT MESSAGE: From '{e.Result.ParticipantId}': '{e.Result.Text}'");
foreach (var entry in e.Result.Translations)
{
Console.WriteLine($"\tTRANSLATED: '{entry.Key}': '{entry.Value}'");
}
};
conversationTranslator.Transcribed += (s, e) =>
{
Console.WriteLine($"TRANSCRIBED: From '{e.Result.ParticipantId}': '{e.Result.Text}'");
foreach (var entry in e.Result.Translations)
{
Console.WriteLine($"\tTRANSLATED: '{entry.Key}': '{entry.Value}'");
}
};
conversationTranslator.Transcribing += (s, e) =>
{
Console.WriteLine($"TRANSCRIBING: From '{e.Result.ParticipantId}': '{e.Result.Text}'");
foreach (var entry in e.Result.Translations)
{
Console.WriteLine($"\tTRANSLATED: '{entry.Key}': '{entry.Value}'");
}
};
// Join the conversation so you can start receiving events
await conversationTranslator.JoinConversationAsync(conversation, displayName).ConfigureAwait(false);
// You can now send an instant message to all other participants in the room
await conversationTranslator.SendTextMessageAsync("The instant message to send").ConfigureAwait(false);
// Start sending audio
await conversationTranslator.StartTranscribingAsync().ConfigureAwait(false);
// At this point, you should start receiving transcriptions for what you are saying using
// the default microphone. Press enter to stop audio capture
Console.WriteLine("Started transcribing. Press enter to stop");
while (Console.ReadKey(true).Key != ConsoleKey.Enter)
{
}
// Stop audio capture
await conversationTranslator.StopTranscribingAsync().ConfigureAwait(false);
// Leave the conversation. After this you will no longer receive events
await conversationTranslator.LeaveConversationAsync().ConfigureAwait(false);
}
// End the conversation
await conversation.EndConversationAsync().ConfigureAwait(false);
// Delete the conversation. Any other participants that are still in the conversation will be removed
await conversation.DeleteConversationAsync().ConfigureAwait(false);
}
}
// Translation from microphone.
public static async Task TranslationWithMicrophoneAsync()
{
// <TranslationWithMicrophoneAsync>
// Translation source language.
// Replace with a language of your choice.
string fromLanguage = "en-US";
// Voice name of synthesis output.
const string GermanVoice = "Microsoft Server Speech Text to Speech Voice (de-DE, Hedda)";
// Creates an instance of a speech translation config with specified subscription key and service region.
// Replace with your own subscription key and service region (e.g., "westus").
var config = SpeechTranslationConfig.FromSubscription("", "westus");
config.SpeechRecognitionLanguage = fromLanguage;
config.VoiceName = GermanVoice;
// Translation target language(s).
// Replace with language(s) of your choice.
config.AddTargetLanguage("de");
// Creates a translation recognizer using microphone as audio input.
using (var recognizer = new TranslationRecognizer(config))
{
// Subscribes to events.
recognizer.Recognizing += (s, e) =>
{
//Console.WriteLine($"RECOGNIZING in '{fromLanguage}': Text={e.Result.Text}");
//foreach (var element in e.Result.Translations)
//{
// Console.WriteLine($" TRANSLATING into '{element.Key}': {element.Value}");
//}
};
recognizer.Recognized += (s, e) =>
{
if (e.Result.Reason == ResultReason.TranslatedSpeech)
{
Console.WriteLine($"\nRECOGNIZED in '{fromLanguage}': Text={e.Result.Text}");
foreach (var element in e.Result.Translations)
{
Console.WriteLine($"TRANSLATED into '{element.Key}': {element.Value}");
}
}
else if (e.Result.Reason == ResultReason.RecognizedSpeech)
{
//Console.WriteLine($"\nText={e.Result.Text}");
//Console.WriteLine($" Speech not translated.");
}
else if (e.Result.Reason == ResultReason.NoMatch)
{
//Console.WriteLine($"NOMATCH: Speech could not be recognized.");
}
};
recognizer.Synthesizing += (s, e) =>
{
//var audio = e.Result.GetAudio();
//Console.WriteLine(audio.Length != 0
// ? $"AudioSize: {audio.Length}"
// : $"AudioSize: {audio.Length} (end of synthesis data)");
//if (audio.Length > 0)
//{
// #if NET461
// using (var m = new MemoryStream(audio))
// {
// SoundPlayer simpleSound = new SoundPlayer(m);
// simpleSound.PlaySync();
// }
// #endif
//}
};
recognizer.Canceled += (s, e) =>
{
Console.WriteLine($"CANCELED: Reason={e.Reason}");
if (e.Reason == CancellationReason.Error)
{
Console.WriteLine($"CANCELED: ErrorCode={e.ErrorCode}");
Console.WriteLine($"CANCELED: ErrorDetails={e.ErrorDetails}");
Console.WriteLine($"CANCELED: Did you update the subscription info?");
}
};
recognizer.SessionStarted += (s, e) =>
{
Console.WriteLine("\nSession started event.");
};
recognizer.SessionStopped += (s, e) =>
{
Console.WriteLine("\nSession stopped event.");
};
// Starts continuous recognition. Uses StopContinuousRecognitionAsync() to stop recognition.
Console.WriteLine("Say something...");
await recognizer.StartContinuousRecognitionAsync().ConfigureAwait(false);
do
{
Console.WriteLine("Press Enter to stop");
} while (Console.ReadKey().Key != ConsoleKey.Enter);
// Stops continuous recognition.
await recognizer.StopContinuousRecognitionAsync().ConfigureAwait(false);
}
// </TranslationWithMicrophoneAsync>
}
public static async Task TranslationContinuousRecognitionAsync(SpeechTranslationConfig config)
{
byte[] audio = null;
string fromLanguage = "en-US";
#region LanguageDetection
/*SpeechConfig speechConfig = SpeechConfig.FromEndpoint(new System.Uri(ConfigurationManager.AppSettings.Get("SpeechEndpoint")), ConfigurationManager.AppSettings.Get("TTSKey"));
* AudioConfig audioConfig = AudioConfig.FromDefaultMicrophoneInput();
* string fromLanguage = string.Empty;
* AutoDetectSourceLanguageConfig autoDetectSourceLanguageConfig = AutoDetectSourceLanguageConfig
* .FromLanguages(new string[] { "en-US", "ru-RU" });
* using (var recognizer = new SpeechRecognizer(
* speechConfig,
* autoDetectSourceLanguageConfig,
* audioConfig))
* {
* Console.WriteLine("Say something...");
* var speechRecognitionResult = await recognizer.RecognizeOnceAsync();
* var autoDetectSourceLanguageResult =
* AutoDetectSourceLanguageResult.FromResult(speechRecognitionResult);
* fromLanguage = autoDetectSourceLanguageResult.Language;
* Console.WriteLine("I recognized " + speechRecognitionResult.Text + " in " + fromLanguage);
* }*/
#endregion
config.SpeechRecognitionLanguage = fromLanguage;
config.AddTargetLanguage("de");
const string GermanVoice = "de-DE-Hedda";
config.VoiceName = GermanVoice;
// Creates a translation recognizer using microphone as audio input.
using (var recognizer = new TranslationRecognizer(config))
{
recognizer.Recognizing += (s, e) =>
{
Console.WriteLine($"RECOGNIZING in '{fromLanguage}': Text={e.Result.Text}");
foreach (var element in e.Result.Translations)
{
Console.WriteLine($" TRANSLATING into '{element.Key}': {element.Value}");
}
};
recognizer.Recognized += (s, e) =>
{
if (e.Result.Reason == ResultReason.TranslatedSpeech)
{
Console.WriteLine($"\nFinal result: Reason: {e.Result.Reason.ToString()}, recognized text in {fromLanguage}: {e.Result.Text}.");
foreach (var element in e.Result.Translations)
{
Console.WriteLine($" TRANSLATING into '{element.Key}': {element.Value}");
}
}
};
recognizer.Synthesizing += (s, e) =>
{
audio = e.Result.GetAudio();
Console.WriteLine(audio.Length != 0
? $"AudioSize: {audio.Length}"
: $"AudioSize: {audio.Length} (end of synthesis data)");
using (MemoryStream ms = new MemoryStream(audio))
{
SoundPlayer player = new SoundPlayer();
player.Stream = null;
player.Stream = ms;
player.Stream.Position = 0;
player.PlaySync();
}
};
recognizer.Canceled += (s, e) =>
{
Console.WriteLine($"\nRecognition canceled. Reason: {e.Reason}; ErrorDetails: {e.ErrorDetails}");
};
recognizer.SessionStarted += (s, e) =>
{
Console.WriteLine("\nSession started event.");
};
recognizer.SessionStopped += (s, e) =>
{
Console.WriteLine("\nSession stopped event.");
};
// Starts continuous recognition. Uses StopContinuousRecognitionAsync() to stop recognition.
Console.WriteLine("Say something...");
await recognizer.RecognizeOnceAsync();//.StartContinuousRecognitionAsync().ConfigureAwait(false);
do
{
Console.WriteLine("Press Enter to stop");
} while (Console.ReadKey().Key != ConsoleKey.Enter);
// Stops continuous recognition.
await recognizer.StopContinuousRecognitionAsync().ConfigureAwait(false);
}
}
