private void Recognized(object sender, SpeechRecognitionEventArgs e)
{
Console.WriteLine();
var test = GetResponse(e.Result.Properties);
SpeechResult.Add(test);
}
public static void main(string[] args)
{
[email protected]("Loading models...");
Configuration configuration = new Configuration();
configuration.setAcousticModelPath("resource:/edu/cmu/sphinx/models/en-us/en-us");
configuration.setDictionaryPath("resource:/edu/cmu/sphinx/models/en-us/cmudict-en-us.dict");
Context context = new Context(configuration);
context.setLocalProperty("decoder->searchManager", "allphoneSearchManager");
Recognizer recognizer = (Recognizer)context.getInstance(ClassLiteral <Recognizer> .Value);
InputStream resourceAsStream = ClassLiteral <AllphoneDemo> .Value.getResourceAsStream("/edu/cmu/sphinx/demo/aligner/10001-90210-01803.wav");
resourceAsStream.skip((long)((ulong)44));
recognizer.allocate();
context.setSpeechSource(resourceAsStream, TimeFrame.INFINITE);
Result result;
while ((result = recognizer.recognize()) != null)
{
SpeechResult speechResult = new SpeechResult(result);
[email protected]("Hypothesis: %s\n", new object[]
{
speechResult.getHypothesis()
});
[email protected]("List of recognized words and their times:");
Iterator iterator = speechResult.getWords().iterator();
while (iterator.hasNext())
{
WordResult wordResult = (WordResult)iterator.next();
[email protected](wordResult);
}
}
recognizer.deallocate();
}
/// <summary>
/// Adds the next recognition result received from the speech recognition service.
/// </summary>
/// <param name="recognitionResult">The recognition result to append.</param>
/// <remarks>
/// The recognition results are accumulated as they are received from the speech service, and
/// may be used at a later time to build the partial and final speech recognition results.
/// </remarks>
internal void AppendResult(SpeechResult recognitionResult)
{
if ((recognitionResult is PartialRecognitionResult || (recognitionResult is RecognitionResult result && result.RecognitionStatus == RecognitionStatus.Success)) &&
this.recognitionResults[this.recognitionResults.Count - 1] is PartialRecognitionResult)
{
// replace the immediately preceding partial result
this.recognitionResults[this.recognitionResults.Count - 1] = recognitionResult;
}
private async void SpeechExecute()
{
SpeechResult result = await this.speechService.RecognizeAsync(null, null);
if (result.IsSuccess)
{
var parameter = this.SelectedFolder.GetTaskCreationParameters();
parameter.Title = result.Text;
this.NavigationService.FlyoutTo(ViewLocator.CreateEditTaskPageNew, parameter);
}
}
private void recognize()
{
waveInStream.StopRecording();
mem.Position = 0;
_recognizer.StartRecognition(mem, new TimeFrame(mem.Length));
SpeechResult result = _recognizer.GetResult();
_recognizer.StopRecognition();
Console.WriteLine("result: " + result.GetHypothesis());
Console.ReadKey();
Console.ReadKey();
}
async void OnRecordReleased(object sender, EventArgs e)
{
DependencyService.Get <IAudioRecorder>().Stop();
btnRecord.Text = "正识别中";
//使用百度API进行语音识别
//OutputText.Text = await ToTextByBaidu();
SpeechResult result = await mmx.Speech.Asr(filepath);
OutputText.Text = result.text;
btnRecord.Text = "录音识别";
}
/// <summary>
/// Function that is called when the Watson API returns a result.
/// </summary>
/// <param name="results">List of speech-to-text results</param>
protected void OnSpeechToTextResult(SpeechResultList results)
{
if (results.HasResult())
{
SpeechResult watsonResult = results.Results[0];
var textResult = m_WatsonSpeechToTextComponent.CreateSpeechToTextResult(watsonResult);
if (m_OnTextResult != null)
{
m_OnTextResult(textResult);
}
m_LastResult = textResult;
}
}
/// <summary>
/// 语音识别，语音转文本
/// </summary>
public static async Task <SpeechResult> Asr(string filepath)
{
SpeechResult result = new SpeechResult();
if (File.Exists(filepath))
{
//var APP_ID = "14965195";
//var API_KEY = "R2qXXgwr9xKtge3kxU5U7up2";
//var SECRET_KEY = "Gnm2KhHcgZEDDLwy0Qtl66y4fFc8FmTj";
var client = new Baidu.Aip.Speech.Asr(APP_ID, API_KEY, SECRET_KEY);
client.Timeout = 60000; // 修改超时时间
//读取文件
//string rootPath = Directory.GetCurrentDirectory();
var data = File.ReadAllBytes(filepath);
//识别语种，英文1737;
Dictionary <string, object> op = new Dictionary <string, object>();
op["dev_pid"] = 1737;
//client.Timeout = 120000; // 若语音较长，建议设置更大的超时时间. ms
var res = Task.Run(() =>
{
var result1 = client.Recognize(data, "amr", 16000, op);
MResult mResult = JsonConvert.DeserializeObject <MResult>(result1.ToString());
if (mResult.err_no == 0)
{
//设置成功返回数据
result.status = 0;
result.text = mResult.result[0].ToString();
}
else
{
//设置失败返回数据
result.status = 1;
result.error = "语音错误：" + mResult.err_no.ToString();
}
});
await res;
}
else
{
//失败数据
result.status = 1;
result.error = "语音错误：无语音";
}
return(result);
}
private void OnOutputReceived(object sender, DataReceivedEventArgs e)
{
try
{
lock (LOCK)
{
result = SpeechResult.Deserialize(e.Data);
VRLog.Info("RECEIVED MESSAGE: " + e.Data);
}
}
catch (Exception err)
{
VRLog.Error(err);
}
}
/// <summary>
/// Populates and returns a SpeechToTextResult object from a given Watson SpeechResult object.
/// </summary>
/// <param name="watsonResult">Watson SpeechResult object</param>
/// <returns>A SpeechToTextResult object</returns>
public SpeechToTextResult CreateSpeechToTextResult(SpeechResult watsonResult)
{
var textResult = new SpeechToTextResult();
textResult.IsFinal = watsonResult.Final;
textResult.TextAlternatives = new TextAlternative[watsonResult.Alternatives.Length];
for (int i = 0; i < textResult.TextAlternatives.Length; ++i)
{
SpeechAlt watsonAlternative = watsonResult.Alternatives[i];
var alternative = new WatsonTextAlternative();
alternative.Text = watsonAlternative.Transcript;
alternative.Confidence = (float)watsonAlternative.Confidence;
alternative.TimeStamps = watsonAlternative.Timestamps;
alternative.WordConfidenceValues = watsonAlternative.WordConfidence;
textResult.TextAlternatives[i] = alternative;
}
return(textResult);
}
private async void QuickSpeechExecute()
{
SpeechResult result = await this.speechService.RecognizeAsync(null, null);
if (result.IsSuccess)
{
var parameter = this.SelectedFolder.GetTaskCreationParameters();
parameter.Title = result.Text;
this.NavigationService.FlyoutTo(ViewLocator.CreateEditTaskPageNew, parameter);
}
else
{
await this.messageBoxService.ShowAsync(
StringResources.Message_Warning,
StringResources.Speech_ErrorDuringRecognitionFormat.TryFormat(result.Text));
}
}
/// <summary>
/// 语音合成，文本转语音
/// </summary>
/// <param name="text"></param>
/// <param name="spd"></param>
/// <param name="pit"></param>
/// <returns></returns>
public static async Task <SpeechResult> Tts(string text, int spd, int pit)
{
SpeechResult result = new SpeechResult();
//var APP_ID = "14965195";
//var API_KEY = "R2qXXgwr9xKtge3kxU5U7up2";
//var SECRET_KEY = "Gnm2KhHcgZEDDLwy0Qtl66y4fFc8FmTj";
var client = new Baidu.Aip.Speech.Tts(API_KEY, SECRET_KEY);
client.Timeout = 60000; // 修改超时时间
// 可选参数
var option = new Dictionary <string, object>()
{
{ "spd", spd }, // 语速
{ "vol", 7 }, // 音量
{ "pit", pit }, // 语调
{ "per", 0 }, // 发音人，0为普通女声，1为普通男生，3为情感合成-度逍遥，4为情感合成-度丫丫，默认为普通女声
{ "aue", 3 }
};
var res = Task.Run(() =>
{
var res1 = client.Synthesis(text, option);
if (res1.ErrorCode == 0) // 或 result.Success
{
//成功数据
result.status = 0;
result.speech = res1.Data;
//File.WriteAllBytes("合成的语音文件本地存储地址.mp3", result.Data);
}
else
{
//失败数据
result.status = 1;
result.error = "错误:" + res1.ErrorCode.ToString();
}
});
await res;
return(result);
}
private async Task SpeechExecute(Action <string> onSuccess, bool isNote)
{
SpeechResult result = await this.speechService.RecognizeAsync(null, null);
if (result.IsSuccess && !string.IsNullOrWhiteSpace(result.Text))
{
onSuccess(result.Text);
if (isNote && this.NoteSpeechRecognitionCompleted != null)
{
this.NoteSpeechRecognitionCompleted(this, new EventArgs <string>(result.Text));
}
}
else
{
await this.messageBoxService.ShowAsync(
StringResources.Message_Warning,
StringResources.Speech_ErrorDuringRecognitionFormat.TryFormat(result.Text));
}
}
async void Record_OnClick()
{
try
{
var audioRecordingService = DependencyService.Get <IAudioRecorderService>();
if (!isRecording)
{
recording.IsVisible = true;
record.IsVisible = false;
audioRecordingService.StartRecording();
// IsProcessing = true;
}
else
{
recording.IsVisible = false;
record.IsVisible = true;
audioRecordingService.StopRecording();
}
isRecording = !isRecording;
if (!isRecording)
{
speechResult = await bingSpeechService.RecognizeSpeechAsync(Constants.AudioFilename);
if (!string.IsNullOrWhiteSpace(speechResult.DisplayText))
{
var result = new SpeechItem();
result.SpeechText = speechResult.DisplayText;
result.Speaker = " - ";
ProcessResponse(result);
}
}
}
catch (Exception ex)
{
return;
//Debug.WriteLine(ex.Message);
}
finally
{
}
}
public void runFreeForm()
{
iSpeechRecognizer iSpeech = new iSpeechRecognizer(_api, _production);
iSpeech.setFreeForm(iSpeechRecognizer.FREEFORM_DICTATION);
StreamingBuffer streaming = iSpeech.startStreamingRecognize("audio/x-wav", this);
uploadFile(_freeFormAudio, streaming);
try
{
SpeechResult result = iSpeech.stopStreaming();
Console.WriteLine("Text = " + result.Text + " Confidence = " + result.Confidence);
}
catch (Exception e)
{
Console.WriteLine("error " + e.ToString());
}
}
public SpeechResult GetIntent(SpeechResult speech, string appId = "db72800d-c433-46fa-83e6-1e07648fa698", string key = "c9baf714974b4a7792324a7c4d3a916e")
{
string html = String.Empty;
var query = HttpUtility.UrlEncode(speech.Lexical);
string requestUri = "https://api.projectoxford.ai/luis/v1/application?subscription-key=" + key + "&id=" + appId + "&q=" + query;
HttpWebRequest request = (HttpWebRequest)WebRequest.Create(requestUri);
request.AutomaticDecompression = DecompressionMethods.GZip;
using (HttpWebResponse response = (HttpWebResponse)request.GetResponse())
using (Stream stream = response.GetResponseStream())
using (StreamReader reader = new StreamReader(stream))
{
html = reader.ReadToEnd();
}
var intentResult = JsonConvert.DeserializeObject <IntentResult>(html);
speech.IntentResult = intentResult;
return(speech);
}
public void runList()
{
iSpeechRecognizer iSpeech = new iSpeechRecognizer(_api, _production);
iSpeech.setFreeForm(iSpeechRecognizer.FREEFORM_DISABLED);
String[] names = new String[] { "jane", "bob", "john" };
iSpeech.addAlias("NAMES", names);
iSpeech.addCommand("call %NAMES%");
StreamingBuffer streaming = iSpeech.startStreamingRecognize("audio/x-wav", this);
uploadFile(_listAudio, streaming);
try
{
SpeechResult result = iSpeech.stopStreaming();
Console.WriteLine("Text = " + result.Text + " Confidence = " + result.Confidence);
}
catch (Exception e)
{
Console.WriteLine("error " + e.ToString());
}
}
private SpeechResultList ParseRecognizeResponse(IDictionary resp)
{
if (resp == null)
{
return(null);
}
try
{
List <SpeechResult> results = new List <SpeechResult>();
IList iresults = resp["results"] as IList;
if (iresults == null)
{
return(null);
}
foreach (var r in iresults)
{
IDictionary iresult = r as IDictionary;
if (iresults == null)
{
continue;
}
SpeechResult result = new SpeechResult();
result.Final = (bool)iresult["final"];
IList ialternatives = iresult["alternatives"] as IList;
if (ialternatives == null)
{
continue;
}
List <SpeechAlt> alternatives = new List <SpeechAlt>();
foreach (var a in ialternatives)
{
IDictionary ialternative = a as IDictionary;
if (ialternative == null)
{
continue;
}
SpeechAlt alternative = new SpeechAlt();
alternative.Transcript = (string)ialternative["transcript"];
if (ialternative.Contains("confidence"))
{
alternative.Confidence = (double)ialternative["confidence"];
}
if (ialternative.Contains("timestamps"))
{
IList itimestamps = ialternative["timestamps"] as IList;
TimeStamp[] timestamps = new TimeStamp[itimestamps.Count];
for (int i = 0; i < itimestamps.Count; ++i)
{
IList itimestamp = itimestamps[i] as IList;
if (itimestamp == null)
{
continue;
}
TimeStamp ts = new TimeStamp();
ts.Word = (string)itimestamp[0];
ts.Start = (double)itimestamp[1];
ts.End = (double)itimestamp[2];
timestamps[i] = ts;
}
alternative.Timestamps = timestamps;
}
if (ialternative.Contains("word_confidence"))
{
IList iconfidence = ialternative["word_confidence"] as IList;
WordConfidence[] confidence = new WordConfidence[iconfidence.Count];
for (int i = 0; i < iconfidence.Count; ++i)
{
IList iwordconf = iconfidence[i] as IList;
if (iwordconf == null)
{
continue;
}
WordConfidence wc = new WordConfidence();
wc.Word = (string)iwordconf[0];
wc.Confidence = (double)iwordconf[1];
confidence[i] = wc;
}
alternative.WordConfidence = confidence;
}
alternatives.Add(alternative);
}
result.Alternatives = alternatives.ToArray();
results.Add(result);
}
return(new SpeechResultList(results.ToArray()));
}
catch (Exception e)
{
Log.Error("SpeechToText", "ParseJsonResponse exception: {0}", e.ToString());
return(null);
}
}
public virtual void collect(SpeechResult result)
{
Token token = result.getResult().getBestToken();
if (token == null)
{
string text = "Best token not found!";
throw new Exception(text);
}
do
{
FloatData floatData = (FloatData)token.getData();
SearchState searchState = token.getSearchState();
if (!(searchState is HMMSearchState) || !searchState.isEmitting())
{
token = token.getPredecessor();
}
else
{
this.nFrames++;
float[] array = token.calculateComponentScore(floatData);
float[] values = FloatData.toFloatData(floatData).getValues();
int num = (int)((HMMSearchState)token.getSearchState()).getHMMState().getMixtureId();
if (this.loader is Sphinx3Loader && this.loader.hasTiedMixtures())
{
num = this.loader.getSenone2Ci()[num];
}
int[] vectorLength = this.loader.getVectorLength();
int numStreams = this.loader.getNumStreams();
int numGaussiansPerState = this.loader.getNumGaussiansPerState();
float[] array2 = this.computePosterios(array, numStreams);
int num2 = 0;
for (int i = 0; i < numStreams; i++)
{
for (int j = 0; j < numGaussiansPerState; j++)
{
int classIndex = this.means.getClassIndex(num * numStreams * numGaussiansPerState + i * numGaussiansPerState + j);
float num3 = array2[i * numGaussiansPerState + j];
if ((double)num3 > (double)0f)
{
float[] array3 = (float[])this.loader.getMeansPool().get(num * numStreams * numGaussiansPerState + i * numGaussiansPerState + j);
for (int k = 0; k < vectorLength[i]; k++)
{
float num4 = array2[i * numGaussiansPerState + j] * values[k + num2];
float num5 = num4 * ((float[])this.loader.getVariancePool().get(num * numStreams * numGaussiansPerState + i * numGaussiansPerState + j))[k];
float num6 = num3 * ((float[])this.loader.getVariancePool().get(num * numStreams * numGaussiansPerState + i * numGaussiansPerState + j))[k];
int num8;
double[] array5;
for (int l = 0; l < vectorLength[i]; l++)
{
float num7 = num6 * array3[l];
for (int m = l; m < vectorLength[i]; m++)
{
double[] array4 = this.regLs[classIndex][i][k][l];
num8 = m;
array5 = array4;
array5[num8] += (double)(num7 * array3[m]);
}
double[] array6 = this.regLs[classIndex][i][k][l];
num8 = vectorLength[i];
array5 = array6;
array5[num8] += (double)num7;
double[] array7 = this.regRs[classIndex][i][k];
num8 = l;
array5 = array7;
array5[num8] += (double)(num5 * array3[l]);
}
double[] array8 = this.regLs[classIndex][i][k][vectorLength[i]];
num8 = vectorLength[i];
array5 = array8;
array5[num8] += (double)num6;
double[] array9 = this.regRs[classIndex][i][k];
num8 = vectorLength[i];
array5 = array9;
array5[num8] += (double)num5;
}
}
}
num2 += vectorLength[i];
}
token = token.getPredecessor();
}
}while (token != null);
}
private async Task <IMessageActivity> GetResponseFromSpeechResultAsync(ITurnContext turnContext,
SpeechResult speechResult)
{
if (speechResult == null)
{
return(null);
}
var reply = turnContext.Activity.CreateReply();
if (speechResult.DisplayText.Contains("도움"))
{
var helpAttachment = await MakeAttachmentAsync("도움말");
reply.Text = "help message";
reply.Attachments.Add(helpAttachment);
}
else if (speechResult.DisplayText.Contains("운전시작") || speechResult.DisplayText.Contains("운전 시작"))
{
var runAttachment = await MakeAttachmentAsync("운전시작");
reply.Text = "run message";
reply.Attachments.Add(runAttachment);
}
else if (speechResult.DisplayText.Contains("운전종료") || speechResult.DisplayText.Contains("운전 종료"))
{
var stopAttachment = await MakeAttachmentAsync("운전종료");
reply.Text = "stop message";
reply.Attachments.Add(stopAttachment);
}
else if (speechResult.DisplayText.Contains("오늘 날씨") || speechResult.DisplayText.Contains("오늘날씨") ||
speechResult.DisplayText.Contains("오늘의날씨"))
{
//날씨 api를 조회해서 결과 반환 - 일자시간으로 파일명 구성해서 저장
//일자시간에 해당하는 내용이 있는지 확인해서 없으면 음성 생성하고 BLOB에 업로드 후 경로 반환
//있으면 해당 음성 파일 경로 반환
var weatherAttachment = await MakeAttachmentAsync("오늘날씨");
reply.Text = "today weather message";
reply.Attachments.Add(weatherAttachment);
}
else if (speechResult.DisplayText.Contains("미세 먼지 정보") || speechResult.DisplayText.Contains("미세먼지 정보") ||
speechResult.DisplayText.Contains("미세먼지정보") || speechResult.DisplayText.Contains("미세먼지") ||
speechResult.DisplayText.Contains("미세 먼지"))
{
//미세먼지 정보를 조회해서 음성 정보로 생성 - 일자시간으로 파일명 구성해서 저장
//일자시간에 해당하는 내용이 있는지 확인해서 없으면 음성 생성하고 BLOB에 업로드 후 경로 반환
//있으면 해당 음성 파일 경로 반환
var dustAttachment = await MakeAttachmentAsync("미세먼지정보");
reply.Text = "today dust message";
reply.Attachments.Add(dustAttachment);
}
else
{
reply.Text = "Unknown command";
reply.Value = speechResult.DisplayText;
}
return(reply);
}
public void Collect(SpeechResult result)
{
Token token = result.Result.GetBestToken();
float[] componentScore, featureVector, posteriors, tmean;
int[] len;
float dnom, wtMeanVar, wtDcountVar, wtDcountVarMean, mean;
int mId, cluster;
int numStreams, gauPerState;
if (token == null)
{
throw new Exception("Best token not found!");
}
do
{
FloatData feature = (FloatData)token.Data;
ISearchState ss = token.SearchState;
if (!(ss is IHMMSearchState && ss.IsEmitting))
{
token = token.Predecessor;
continue;
}
componentScore = token.CalculateComponentScore(feature);
featureVector = FloatData.ToFloatData(feature).Values;
mId = (int)((IHMMSearchState)token.SearchState).HmmState
.GetMixtureId();
if (_loader is Sphinx3Loader && _loader.HasTiedMixtures())
{
// use CI phone ID for tied mixture model
mId = _loader.Senone2Ci[mId];
}
len = _loader.VectorLength;
numStreams = _loader.NumStreams;
gauPerState = _loader.NumGaussiansPerState;
posteriors = ComputePosterios(componentScore, numStreams);
int featVectorStartIdx = 0;
for (int i = 0; i < numStreams; i++)
{
for (int j = 0; j < gauPerState; j++)
{
cluster = _means.GetClassIndex(mId * numStreams
* gauPerState + i * gauPerState + j);
dnom = posteriors[i * gauPerState + j];
if (dnom > 0f)
{
tmean = _loader.MeansPool.Get(
mId * numStreams * gauPerState + i
* gauPerState + j);
for (int k = 0; k < len[i]; k++)
{
mean = posteriors[i * gauPerState + j]
* featureVector[k + featVectorStartIdx];
wtMeanVar = mean
* _loader.VariancePool.Get(
mId * numStreams * gauPerState + i
* gauPerState + j)[k];
wtDcountVar = dnom
* _loader.VariancePool.Get(
mId * numStreams * gauPerState + i
* gauPerState + j)[k];
for (int p = 0; p < len[i]; p++)
{
wtDcountVarMean = wtDcountVar * tmean[p];
for (int q = p; q < len[i]; q++)
{
RegLs[cluster][i][k][p][q] += wtDcountVarMean
* tmean[q];
}
RegLs[cluster][i][k][p][len[i]] += wtDcountVarMean;
RegRs[cluster][i][k][p] += wtMeanVar * tmean[p];
}
RegLs[cluster][i][k][len[i]][len[i]] += wtDcountVar;
RegRs[cluster][i][k][len[i]] += wtMeanVar;
}
}
}
featVectorStartIdx += len[i];
}
token = token.Predecessor;
} while (token != null);
}
/// <summary>
/// Converts a <see cref="SpeechResult"/> to a <see cref="StreamingSpeechRecognitionResult"/>.
/// </summary>
/// <param name="result">The <see cref="SpeechResult"/>.</param>
/// <param name="rankOrder">A value indicating whether to score the alternates by their rank-order.</param>
/// <returns>The <see cref="StreamingSpeechRecognitionResult"/>.</returns>
internal static StreamingSpeechRecognitionResult ToStreamingSpeechRecognitionResult(this SpeechResult result, bool rankOrder = true)
{
if (result is PartialRecognitionResult partialResult)
{
return(new StreamingSpeechRecognitionResult(
false,
partialResult.Text,
1.0,
new SpeechRecognitionAlternate[] { new SpeechRecognitionAlternate(partialResult.Text, 1.0) },
null,
result.Duration));
}
else if (result is RecognitionResult recoResult)
{
var alternates = recoResult.Results?.Select(r => new SpeechRecognitionAlternate(r.LexicalForm, r.Confidence)) ?? new[] { new SpeechRecognitionAlternate(string.Empty, 1.0) };
if (!rankOrder)
{
// order by confidence score
alternates = alternates.OrderByDescending(r => r.Confidence);
}
var topAlternate = alternates.First();
return(new StreamingSpeechRecognitionResult(
false,
topAlternate.Text,
topAlternate.Confidence,
alternates,
null,
result.Duration));
}
else
{
throw new InvalidOperationException("Unexpected recognition result type!");
}
}
public SpeechRecognizedEventArgs(SpeechResult result)
{
Result = result;
}
public void collect(SpeechResult result)
{
Token token = result.getResult().getBestToken();
float[] componentScore, featureVector, posteriors, tmean;
int[] len;
float dnom, wtMeanVar, wtDcountVar, wtDcountVarMean, mean;
int mId, cluster;
int numStreams, gauPerState;
if (token == null)
{
throw new Exception("Best token not found!");
}
do
{
FloatData feature = (FloatData)token.getData();
ISearchState ss = token.getSearchState();
if (!(ss is IHMMSearchState && ss.isEmitting()))
{
token = token.getPredecessor();
continue;
}
componentScore = token.calculateComponentScore(feature);
featureVector = FloatData.toFloatData(feature).getValues();
mId = (int)((IHMMSearchState)token.getSearchState()).getHMMState()
.getMixtureId();
if (loader is Sphinx3Loader && ((Sphinx3Loader)loader).hasTiedMixtures())
{
// use CI phone ID for tied mixture model
mId = ((Sphinx3Loader)loader).getSenone2Ci()[mId];
}
len = loader.getVectorLength();
numStreams = loader.getNumStreams();
gauPerState = loader.getNumGaussiansPerState();
posteriors = this.computePosterios(componentScore, numStreams);
int featVectorStartIdx = 0;
for (int i = 0; i < numStreams; i++)
{
for (int j = 0; j < gauPerState; j++)
{
cluster = means.getClassIndex(mId * numStreams
* gauPerState + i * gauPerState + j);
dnom = posteriors[i * gauPerState + j];
if (dnom > 0f)
{
tmean = loader.getMeansPool().get(
mId * numStreams * gauPerState + i
* gauPerState + j);
for (int k = 0; k < len[i]; k++)
{
mean = posteriors[i * gauPerState + j]
* featureVector[k + featVectorStartIdx];
wtMeanVar = mean
* loader.getVariancePool().get(
mId * numStreams * gauPerState + i
* gauPerState + j)[k];
wtDcountVar = dnom
* loader.getVariancePool().get(
mId * numStreams * gauPerState + i
* gauPerState + j)[k];
for (int p = 0; p < len[i]; p++)
{
wtDcountVarMean = wtDcountVar * tmean[p];
for (int q = p; q < len[i]; q++)
{
regLs[cluster][i][k][p][q] += wtDcountVarMean
* tmean[q];
}
regLs[cluster][i][k][p][len[i]] += wtDcountVarMean;
regRs[cluster][i][k][p] += wtMeanVar * tmean[p];
}
regLs[cluster][i][k][len[i]][len[i]] += wtDcountVar;
regRs[cluster][i][k][len[i]] += wtMeanVar;
}
}
}
featVectorStartIdx += len[i];
}
token = token.getPredecessor();
} while (token != null);
}
/// <summary>
/// API获取语音，文本转语音
/// </summary>
/// <param name="text">需要转换语音的文本</param>
/// <param name="vol">音量</param>
/// <param name="per">发间人</param>
/// <param name="spd">语速</param>
/// <param name="pit">语调</param>
/// <returns></returns>
public static async Task <SpeechResult> TextToSpeech1(string text, string vol, string per, string spd, string pit, string token)
{
SpeechResult result = new SpeechResult();
string url = "http://tsn.baidu.com/text2audio"; //百度tts请求地址
HttpClient client = new HttpClient();
client.Timeout = new TimeSpan(0, 1, 0); //超时时间设置aue
Dictionary <string, string> param = new Dictionary <string, string>()
{
{ "lan", "zh" },
{ "ctp", "1" },
{ "vol", vol }, //音量：0-15，默认5中音量
{ "per", per }, //发音人：0为普通女声，1为普通男声，3为情感合成-度逍遥，4为情感合成-度丫丫
{ "spd", spd }, //语速：0-15，5为中语速
{ "pit", pit }, //音调：0-15，5为中语调
{ "aue", "3" } //不需要修改的参数
};
param.Add("tex", text); //需要转换的文本内容
param.Add("cuid", "delafqmisspeed"); //用户唯一标识"24.021dc614e13e2f98f43b661fb40495d5.2592000.1557308095.282335-14965195"
param.Add("tok", token); //access_token，在发送之前先调用一次获取到这个值
FormUrlEncodedContent content = new FormUrlEncodedContent(param); //post请求参数设置对象
try
{
HttpResponseMessage x = await client.PostAsync(url, content);
if (x.StatusCode == HttpStatusCode.OK)
{
#region 错误处理
HttpContentHeaders header = x.Content.Headers;
//如果返回错误信息
if (header.ContentType.ToString() == "application/json")
{
string res = await x.Content.ReadAsStringAsync();
var Item = JsonConvert.DeserializeObject <TResult>(res);
//设置失败返回数据
result.status = 1;
result.error = Item.err_no.ToString();
}
#endregion
var resbyte = await x.Content.ReadAsByteArrayAsync();
//设置成功返回数据
result.status = 0;
result.speech = resbyte;
}
else
{
//失败
result.status = 1;
result.error = "请求失败";
}
}
catch (Exception ex)
{
//失败
result.status = 1;
result.error = ex.Message;
}
return(result);
}
public async Task <ActionResult> GetTextToSpeech(string text, int no)
{
if (text == null)
{
text = "Kein text angekommen.";
}
Debug.WriteLine(no);
SpeechResult speechResult = new SpeechResult();
speechResult.TextPartNo = no;
speechResult.Text = text;
#region Bing Speech Authentication
// Note: The way to get api key:
// Free: https://www.microsoft.com/cognitive-services/en-us/subscriptions?productId=/products/Bing.Speech.Preview
// Paid: https://portal.azure.com/#create/Microsoft.CognitiveServices/apitype/Bing.Speech/pricingtier/S0
string bingApiKey = ConfigurationManager.AppSettings["BingSpeechApiKey"];
Authentication auth = new Authentication(bingApiKey);
string accessToken;
try
{
accessToken = auth.GetAccessToken();
speechResult.AccessToken = accessToken;
}
catch (Exception ex)
{
speechResult.ErrorMessage = "GetTextToSpeech --> Failed authentication --> " + ex.Message + " " + ex.ToString();
JsonResult result = new JsonResult()
{
Data = speechResult
};
return(result);
}
#endregion
#region Text-To-Speech
// Debug.WriteLine("Starting TTSSample request code execution.");
string requestUri = "https://speech.platform.bing.com/synthesize";
var cortana = new Synthesize();
cortana.OnAudioAvailable += (s, args) => {
var input = args.EventData;
byte[] buffer = new byte[20 * 1024 * 1024];
using (MemoryStream ms = new MemoryStream())
{
int read;
while ((read = input.Read(buffer, 0, buffer.Length)) > 0)
{
ms.Write(buffer, 0, read);
}
speechResult.SpeechEventData = ms.ToArray();;
}
};
//cortana.OnError += ErrorHandler;
cortana.OnError += (s, e) =>
{
var message = $"Unable to complete the TTS request: {e.ToString()}";
message += e.EventData.Message;
speechResult.ErrorMessage = message;
};
// Reuse Synthesize object to minimize latency
await cortana.Speak(CancellationToken.None,
new Synthesize.InputOptions()
{
RequestUri = new Uri(requestUri),
// Text to be spoken.
Text = text,
VoiceType = Gender.Female,
// Refer to the documentation for complete list of supported locales.
Locale = ConfigurationManager.AppSettings["DefaultLocale"], // "de-DE",
// You can also customize the output voice. Refer to the documentation to view the different
// voices that the TTS service can output.
VoiceName = ConfigurationManager.AppSettings["VoiceName"], // "Microsoft Server Speech Text to Speech Voice (de-De, Hedda)",
// Service can return audio in different output format.
OutputFormat = AudioOutputFormat.Riff16Khz16BitMonoPcm,
AuthorizationToken = "Bearer " + accessToken,
});
JsonResult jsonResult = new JsonResult()
{
Data = speechResult
};
#endregion
return(jsonResult);
}