/// <summary>
/// 创建语音识别的客户端类型的实例。
/// 该实例可以识别来自文件和语音设备的语音。
/// 语音数据会被切分很小的段，然后使用该实例连续的向服务端发送一段。
/// </summary>
/// <param name="mode"><see cref="SpeechRecognitionMode"/>指明语音识别的模式。</param>
private void CreateDataRecoClient(SpeechRecognitionMode mode)
{
if (this.dataClient != null)
{
this.dataClient.Dispose();
this.dataClient = null;
}
// 使用工厂类型的 CreateDataClient 方法创建 DataRecognitionClient 类型的实例。
this.dataClient = SpeechRecognitionServiceFactory.CreateDataClient(
mode, // 指定语音识别的模式。
"en-US", // 我们把语音中语言的类型 hardcode 为英语，因为我们的两个 demo 文件都是英语语音。
SUBSCRIPTIONKEY); // Bing Speech API 服务实例的 key。
// 为语音识别Event handlers for speech recognition results
if (mode == SpeechRecognitionMode.ShortPhrase)
{
// 为 ShortPhrase 模式的识别结果添加处理程序。
this.dataClient.OnResponseReceived += this.OnDataShortPhraseResponseReceivedHandler;
}
else
{
// 为 LongDictation 模式的识别结果添加处理程序。
// 服务端根据分辨出的语句间的停顿会多次触发执行该处理程序。
this.dataClient.OnResponseReceived += this.OnDataDictationResponseReceivedHandler;
}
// 在服务端执行语音识别的过程中，该处理程序会被执行多次，
// 具体是在语音服务对语音的内容产生了预测的结果时，就会触发执行该处理程序。
this.dataClient.OnPartialResponseReceived += this.OnPartialResponseReceivedHandler;
// 在服务端检测到错误时，触发执行该处理程序。
this.dataClient.OnConversationError += this.OnConversationErrorHandler;
}
/// <summary>
/// Initializes a new instance of the <see cref="SpeechRecognitionClient"/> class.
/// </summary>
/// <param name="recognitionMode">The recognition mode to use.</param>
/// <param name="language">The recognition language.</param>
/// <param name="subscriptionKey">The speech service API key.</param>
public SpeechRecognitionClient(SpeechRecognitionMode recognitionMode, string language, string subscriptionKey)
{
this.webSocket = new ClientWebSocket();
this.authentication = new Authentication(subscriptionKey);
this.subscriptionKey = subscriptionKey;
this.language = language;
this.recognitionMode = recognitionMode;
}
/// <summary>
/// using client library=>yet to implement
/// </summary>
/// <param name="mode"></param>
/// <param name="language"></param>
/// <param name="subscriptionKey"></param>
public SpeechToTextService(SpeechRecognitionMode mode, string language, string subscriptionKey)
{
serviceClient = SpeechRecognitionServiceFactory.CreateDataClient(mode, language, subscriptionKey);
// Event handlers for speech recognition results
serviceClient.OnResponseReceived += this.OnResponseReceivedHandler;
serviceClient.OnPartialResponseReceived += this.OnPartialResponseReceivedHandler;
serviceClient.OnConversationError += this.OnConversationErrorHandler;
}
public static void ConvertSpeechToText(SpeechRecognitionMode mode, string language, string subscriptionKey)
{
_microRecogClient = SpeechRecognitionServiceFactory.CreateMicrophoneClient(mode, language, subscriptionKey);
_microRecogClient.OnResponseReceived += OnResponseReceivedHandler;
_microRecogClient.OnPartialResponseReceived += OnPartialResponseReceivedHandler;
_microRecogClient.OnConversationError += OnConversationError;
_microRecogClient.StartMicAndRecognition();
}
async Task ISpeechEngine.SetRecognitionModeAsync(SpeechRecognitionMode mode)
{
if (mode != RecognitionMode)
{
RecognitionMode = mode;
await
(
mode == SpeechRecognitionMode.Paused
? _.EndRecognitionSessionAsync()
: _.StartContinuousRecognitionAsync()
);
}
}
public async Task SetRecognitionModeAsync(SpeechRecognitionMode mode)
{
if (mode != RecognitionMode)
{
RecognitionMode = mode;
await
(
mode == SpeechRecognitionMode.Paused
? EndRecognitionSessionAsync()
: StartContinuousRecognitionAsync()
);
}
}
/// <summary>
/// Assigns the kind of speech that the <see cref="SpeechManager"/> listens
/// for: commands or dictation.
/// </summary>
/// <param name="mode">The recognition mode.</param>
/// <returns>Void</returns>
public async Task SetRecognitionMode(SpeechRecognitionMode mode)
{
if (mode != RecognitionMode)
{
RecognitionMode = mode;
if (mode == SpeechRecognitionMode.Paused)
{
await EndRecognitionSession();
}
else
{
await StartContinuousRecognition();
}
}
}
/// <summary>
/// Initializes a new instance of the <see cref="SpeechRecognitionClient"/> class.
/// </summary>
/// <param name="recognitionMode">The recognition mode to use.</param>
/// <param name="language">The recognition language.</param>
/// <param name="subscriptionKey">The speech service API key.</param>
/// <param name="region">The speech service region associated to the subscription.</param>
public SpeechRecognitionClient(SpeechRecognitionMode recognitionMode, string language, string subscriptionKey, string region = null)
{
this.webSocket = new ClientWebSocket();
this.region = region?.Replace(" ", string.Empty); // convert to region endpoint URL string
if (string.IsNullOrWhiteSpace(this.region))
{
this.authentication = new Authentication(subscriptionKey);
}
else
{
this.authentication = new AzureAuthentication(subscriptionKey, this.region);
}
this.subscriptionKey = subscriptionKey;
this.language = language;
this.recognitionMode = recognitionMode;
}
MicrophoneRecognitionClient CreateMicrophoneRecoClient(string recoLanguage, SpeechRecognitionMode recoMode)
{
this.micClient = SpeechRecognitionServiceFactory.CreateMicrophoneClient(
recoMode,
recoLanguage,
_speechAPIAccountKey);
// Event handlers for speech recognition results
this.micClient.OnMicrophoneStatus += this.OnMicrophoneStatus;
this.micClient.OnPartialResponseReceived += this.OnPartialResponseReceivedHandler;
if (recoMode == SpeechRecognitionMode.ShortPhrase)
{
this.micClient.OnResponseReceived += this.OnMicShortPhraseResponseReceivedHandler;
}
this.micClient.OnConversationError += this.OnConversationErrorHandler;
return(micClient);
}
MicrophoneRecognitionClient CreateMicrophoneRecoClient(SpeechRecognitionMode recoMode, string language, string subscriptionKey)
{
MicrophoneRecognitionClient micClient = SpeechRecognitionServiceFactory.CreateMicrophoneClient(
recoMode,
language,
subscriptionKey);
// Event handlers for speech recognition results
micClient.OnMicrophoneStatus += OnMicrophoneStatus;
micClient.OnPartialResponseReceived += OnPartialResponseReceivedHandler;
micClient.OnResponseReceived += OnMicShortPhraseResponseReceivedHandler;
micClient.OnConversationError += OnConversationErrorHandler;
return(micClient);
}
public void Start(bool isShortParse = true)
{
if (isShortParse)
{
this.Mode = SpeechRecognitionMode.ShortPhrase;
}
else
{
this.Mode = SpeechRecognitionMode.LongDictation;
}
if (this.micClient == null)
{
this.CreateMicrophoneRecoClient();
}
this.micClient.StartMicAndRecognition();
CreateDataRecoClient();
}
/// <summary>
/// Speech recognition with data (for example from a file or audio source).
/// The data is broken up into buffers and each buffer is sent to the Speech Recognition Service.
/// No modification is done to the buffers, so the user can apply their
/// own Silence Detection if desired.
/// </summary>
DataRecognitionClient CreateDataRecoClient(SpeechRecognitionMode recoMode, string recoLanguage)
{
DataRecognitionClient dataClient = SpeechRecognitionServiceFactory.CreateDataClient(
recoMode,
recoLanguage,
SubscriptionKey);
// Event handlers for speech recognition results
if (recoMode == SpeechRecognitionMode.ShortPhrase)
{
dataClient.OnResponseReceived += OnDataShortPhraseResponseReceivedHandler;
}
else
{
dataClient.OnResponseReceived += OnDataDictationResponseReceivedHandler;
}
dataClient.OnPartialResponseReceived += OnPartialResponseReceivedHandler;
dataClient.OnConversationError += OnConversationErrorHandler;
return(dataClient);
}
/// <summary>
/// Assigns the kind of speech that the <see cref="SpeechManager"/> listens
/// for: commands or dictation.
/// </summary>
/// <param name="mode">The recognition mode.</param>
/// <returns>Void</returns>
public async Task <bool> SetRecognitionMode(SpeechRecognitionMode mode)
{
bool success = true;
if (mode != RecognitionMode)
{
RecognitionMode = mode;
if (mode == SpeechRecognitionMode.Paused)
{
await EndRecognitionSession();
}
else
{
if (!await StartContinuousRecognition())
{
success = false;
mode = SpeechRecognitionMode.Default;
}
}
OnModeChanged(new EventArgs());
}
return(success);
}
private void LogRecognitionStart(string recoSource, string recoLanguage, SpeechRecognitionMode recoMode)
{
WriteLine("\n--- Start speech recognition using " + recoSource + " with " + recoMode + " mode in " + recoLanguage + " language ----\n\n");
}
