public async Task SynthesizeSpeechAsync()
{
Mock <TextToSpeech.TextToSpeechClient> mockGrpcClient = new Mock <TextToSpeech.TextToSpeechClient>(MockBehavior.Strict);
SynthesizeSpeechRequest expectedRequest = new SynthesizeSpeechRequest
{
Input = new SynthesisInput(),
Voice = new VoiceSelectionParams(),
AudioConfig = new AudioConfig(),
};
SynthesizeSpeechResponse expectedResponse = new SynthesizeSpeechResponse
{
AudioContent = ByteString.CopyFromUtf8("16"),
};
mockGrpcClient.Setup(x => x.SynthesizeSpeechAsync(expectedRequest, It.IsAny <CallOptions>()))
.Returns(new Grpc.Core.AsyncUnaryCall <SynthesizeSpeechResponse>(Task.FromResult(expectedResponse), null, null, null, null));
TextToSpeechClient client = new TextToSpeechClientImpl(mockGrpcClient.Object, null);
SynthesisInput input = new SynthesisInput();
VoiceSelectionParams voice = new VoiceSelectionParams();
AudioConfig audioConfig = new AudioConfig();
SynthesizeSpeechResponse response = await client.SynthesizeSpeechAsync(input, voice, audioConfig);
Assert.Same(expectedResponse, response);
mockGrpcClient.VerifyAll();
}
private Stream GetAudioStream(string message)
{
SynthesisInput input = new SynthesisInput
{
Text = message
};
VoiceSelectionParams voiceSelection = new VoiceSelectionParams
{
LanguageCode = "es-ES",
Name = "es-ES-Standard-A"
};
AudioConfig audioConfig = new AudioConfig
{
AudioEncoding = AudioEncoding.Linear16,
SpeakingRate = Configuration.BasicConfiguration.MessageSpeed
};
try
{
SynthesizeSpeechResponse response = Client.SynthesizeSpeech(input, voiceSelection, audioConfig);
return(new MemoryStream(response.AudioContent.ToByteArray()));
}
catch (Exception ex)
{
TwitchBot.Instance.LogMessage(LogSeverity.Error, ex.Message, ex.StackTrace);
return(null);
}
}
public static int Main(string[] args)
{
// Create client
TextToSpeechClient client = TextToSpeechClient.Create();
// Initialize request argument(s)
SynthesisInput input = new SynthesisInput
{
Text = "test",
};
VoiceSelectionParams voice = new VoiceSelectionParams
{
LanguageCode = "en-US",
};
AudioConfig audioConfig = new AudioConfig
{
AudioEncoding = AudioEncoding.Mp3,
};
// Call API method
SynthesizeSpeechResponse response = client.SynthesizeSpeech(input, voice, audioConfig);
// Show the result
Console.WriteLine(response);
// Success
Console.WriteLine("Smoke test passed OK");
return(0);
}
public async Task <string> Utter(string text, string dir, VoiceId voice)
{
string speaker = "Chatbot";
ConsoleLogger.WriteLine(speaker, $"{text}");
string fileName = (text + "-" + voice).GetMd5Hash() + ".mp3";
string recordBasePath = $"wwwroot{Path.DirectorySeparatorChar}voice";
string filePath = $"{dir}{Path.DirectorySeparatorChar}{recordBasePath}{Path.DirectorySeparatorChar}{fileName}";
if (File.Exists(filePath))
{
return($"/voice/{fileName}");
}
SynthesizeSpeechRequest sreq = new SynthesizeSpeechRequest();
sreq.Text = text;
sreq.OutputFormat = OutputFormat.Mp3;
sreq.VoiceId = voice;
SynthesizeSpeechResponse sres = await polly.SynthesizeSpeechAsync(sreq);
using (FileStream fileStream = File.Create(filePath))
{
sres.AudioStream.CopyTo(fileStream);
fileStream.Flush();
fileStream.Close();
}
return($"/voice/{fileName}");
}
public static string Speak(string txt, QuestToSpeech.Voice voice, string filePath)
{
try {
TextToSpeechClient client = TextToSpeechClient.Create();
SynthesizeSpeechResponse res = client.SynthesizeSpeech(new SynthesizeSpeechRequest {
Input = new SynthesisInput {
Text = txt
},
Voice = new VoiceSelectionParams {
Name = voice.Name,
LanguageCode = voice.LangCode,
SsmlGender = voice.Gender == QuestToSpeech.Gender.Female ? SsmlVoiceGender.Female : SsmlVoiceGender.Male
},
AudioConfig = new AudioConfig {
AudioEncoding = AudioEncoding.Linear16
}
});
using (FileStream output = File.Create(filePath)) {
res.AudioContent.WriteTo(output);
}
} catch (Exception ex) {
return(string.Format("GoogleTTS Exception: {0}", ex.InnerException == null ? ex.Message : ex.InnerException.ToString()));
}
return(null);
}
public static void SynthesizeSpeech()
{
AmazonPollyClient client = new AmazonPollyClient();
String outputFileName = "speech.mp3";
SynthesizeSpeechRequest synthesizeSpeechRequest = new SynthesizeSpeechRequest()
{
OutputFormat = OutputFormat.Mp3,
VoiceId = VoiceId.Joanna,
Text = "This is a sample text to be synthesized."
};
try
{
using (FileStream outputStream = new FileStream(outputFileName, FileMode.Create, FileAccess.Write))
{
SynthesizeSpeechResponse synthesizeSpeechResponse = client.SynthesizeSpeech(synthesizeSpeechRequest);
byte[] buffer = new byte[2 * 1024];
int readBytes;
Stream inputStream = synthesizeSpeechResponse.AudioStream;
while ((readBytes = inputStream.Read(buffer, 0, 2 * 1024)) > 0)
{
outputStream.Write(buffer, 0, readBytes);
}
}
}
catch (Exception e)
{
Console.WriteLine("Exception caught: " + e.Message);
}
}
public static void TestPolly2()
{
using (AmazonPollyClient pc = new AmazonPollyClient())
{
SynthesizeSpeechRequest sreq = new SynthesizeSpeechRequest();
sreq.Text = "Your Sample Text Here 123";
sreq.OutputFormat = OutputFormat.Mp3;
sreq.VoiceId = VoiceId.Salli;
SynthesizeSpeechResponse sres = pc.SynthesizeSpeechAsync(sreq).GetAwaiter().GetResult();
using (var memoryStream = new MemoryStream())
{
sres.AudioStream.CopyTo(memoryStream);
memoryStream.Flush();
memoryStream.Position = 0;
string outputFile = @".\TestAudio\output-from-polly.wav";
using (Mp3FileReader reader = new Mp3FileReader(memoryStream))
{
using (WaveStream pcmStream = WaveFormatConversionStream.CreatePcmStream(reader))
{
WaveFileWriter.CreateWaveFile(outputFile, pcmStream);
}
}
}
}
}
bool SynthesizeSpeech(string filename, string line, string voice, List <string> lexiconNames)
{
try
{
SynthesizeSpeechRequest sreq = new SynthesizeSpeechRequest();
sreq.Text = line;
sreq.OutputFormat = _fileFormat;
sreq.VoiceId = voice;
if (lexiconNames != null)
{
sreq.LexiconNames = lexiconNames;
}
if (checkBoxSSML.Checked)
{
sreq.TextType = "SSML";
}
SynthesizeSpeechResponse sres = _pc.SynthesizeSpeech(sreq);
using (var fileStream = File.Create(_outputFolder + @"\" + @filename))
{
sres.AudioStream.CopyTo(fileStream);
fileStream.Flush();
fileStream.Close();
}
}
catch (Exception ex)
{
MessageBox.Show("Error in: " + filename + "\n" + ex.Message, "Amazon Error");
return(false);
}
return(true);
}
public async stt::Task SynthesizeSpeechAsync()
{
moq::Mock <TextToSpeech.TextToSpeechClient> mockGrpcClient = new moq::Mock <TextToSpeech.TextToSpeechClient>(moq::MockBehavior.Strict);
SynthesizeSpeechRequest request = new SynthesizeSpeechRequest
{
Input = new SynthesisInput(),
Voice = new VoiceSelectionParams(),
AudioConfig = new AudioConfig(),
};
SynthesizeSpeechResponse expectedResponse = new SynthesizeSpeechResponse
{
AudioContent = proto::ByteString.CopyFromUtf8("audio_content20992f23"),
Timepoints = { new Timepoint(), },
AudioConfig = new AudioConfig(),
};
mockGrpcClient.Setup(x => x.SynthesizeSpeechAsync(request, moq::It.IsAny <grpccore::CallOptions>())).Returns(new grpccore::AsyncUnaryCall <SynthesizeSpeechResponse>(stt::Task.FromResult(expectedResponse), null, null, null, null));
TextToSpeechClient client = new TextToSpeechClientImpl(mockGrpcClient.Object, null);
SynthesizeSpeechResponse responseCallSettings = await client.SynthesizeSpeechAsync(request.Input, request.Voice, request.AudioConfig, gaxgrpc::CallSettings.FromCancellationToken(st::CancellationToken.None));
xunit::Assert.Same(expectedResponse, responseCallSettings);
SynthesizeSpeechResponse responseCancellationToken = await client.SynthesizeSpeechAsync(request.Input, request.Voice, request.AudioConfig, st::CancellationToken.None);
xunit::Assert.Same(expectedResponse, responseCancellationToken);
mockGrpcClient.VerifyAll();
}
async public void getAudio(String text, String path)
{
AmazonPollyClient pc = new AmazonPollyClient(Amazon.RegionEndpoint.USEast1);
SynthesizeSpeechRequest sreq = new SynthesizeSpeechRequest();
sreq.Text = text;
sreq.OutputFormat = OutputFormat.Mp3;
sreq.VoiceId = VoiceId.Vitoria;
sreq.SampleRate = "8000";
SynthesizeSpeechResponse sres = pc.SynthesizeSpeech(sreq);
FileStream fileStream = File.Create(path);
sres.AudioStream.CopyTo(fileStream);
fileStream.Flush();
fileStream.Close();
FileStream fs = File.OpenRead(path);
using (Mp3FileReader reader = new Mp3FileReader(fs))
{
var newFormat = new WaveFormat(8000, 16, 1);
using (var converter = new WaveFormatConversionStream(newFormat, reader))
{
var pathw = path + ".wav";
WaveFileWriter.CreateWaveFile(pathw, converter);
}
}
fs.Dispose();
File.Delete(path);
}
public void SynthesizeSpeechRequestObject()
{
moq::Mock <TextToSpeech.TextToSpeechClient> mockGrpcClient = new moq::Mock <TextToSpeech.TextToSpeechClient>(moq::MockBehavior.Strict);
SynthesizeSpeechRequest request = new SynthesizeSpeechRequest
{
Input = new SynthesisInput(),
Voice = new VoiceSelectionParams(),
AudioConfig = new AudioConfig(),
EnableTimePointing =
{
SynthesizeSpeechRequest.Types.TimepointType.Unspecified,
},
};
SynthesizeSpeechResponse expectedResponse = new SynthesizeSpeechResponse
{
AudioContent = proto::ByteString.CopyFromUtf8("audio_content20992f23"),
Timepoints = { new Timepoint(), },
AudioConfig = new AudioConfig(),
};
mockGrpcClient.Setup(x => x.SynthesizeSpeech(request, moq::It.IsAny <grpccore::CallOptions>())).Returns(expectedResponse);
TextToSpeechClient client = new TextToSpeechClientImpl(mockGrpcClient.Object, null);
SynthesizeSpeechResponse response = client.SynthesizeSpeech(request);
xunit::Assert.Same(expectedResponse, response);
mockGrpcClient.VerifyAll();
}
public void SimpleExample()
{
// Sample: SynthesizeSpeech
// Additional: SynthesizeSpeech(SynthesisInput,VoiceSelectionParams,AudioConfig,CallSettings)
TextToSpeechClient client = TextToSpeechClient.Create();
// The input can be provided as text or SSML.
SynthesisInput input = new SynthesisInput
{
Text = "This is a demonstration of the Google Cloud Text-to-Speech API"
};
// You can specify a particular voice, or ask the server to pick based
// on specified criteria.
VoiceSelectionParams voiceSelection = new VoiceSelectionParams
{
LanguageCode = "en-US",
SsmlGender = SsmlVoiceGender.Female
};
// The audio configuration determines the output format and speaking rate.
AudioConfig audioConfig = new AudioConfig
{
AudioEncoding = AudioEncoding.Mp3
};
SynthesizeSpeechResponse response = client.SynthesizeSpeech(input, voiceSelection, audioConfig);
using (Stream output = File.Create("sample.mp3"))
{
// response.AudioContent is a ByteString. This can easily be converted into
// a byte array or written to a stream.
response.AudioContent.WriteTo(output);
}
// End sample
}
private void OnlineRequestPolly(SynthesizeSpeechRequest sreq, string name)
{
SynthesizeSpeechResponse sres = apc.SynthesizeSpeech(sreq);
using (FileStream fileStream = File.Create("Assets/Resources/" + ResourcePathConstants.SpeechCacheFolder + name + ".mp3")) {
sres.AudioStream.CopyTo(fileStream);
fileStream.Flush();
fileStream.Close();
Debug.Log("New speech file created succesfully");
}
}
private static void t2sOut()
{
while (runSpeech)
{
if (currentText == lastText)
{
continue;
}
if (currentText.StartsWith("+"))
{
continue;
}
currentText = currentText.Replace('^', ' ').Replace(Environment.NewLine, " ").Replace("$s", "").Replace("$h", "").Replace("$g", "").Replace("$e", "").Replace("$u", "").Replace("$b", "").Replace("$8", "").Replace("$l", "").Replace("$q", "").Replace("$9", "").Replace("$a", "").Replace("$7", "").Replace("<", "").Replace("$r", "");
/*
* if (Game1.player.isMarried())
* currentText = currentText.Replace(" " + Game1.player.spouse + " ", " your spouse ").Replace(" " + Game1.player.spouse, " your spouse").Replace(Game1.player.spouse + " ", "Your spouse ");
*
* currentText.Replace(" " + Game1.player.Name + " ", " Farmer ").Replace(" " + Game1.player.Name, " Farmer").Replace(Game1.player.Name + " ", "Farmer ");
*/
int hash = currentText.GetHashCode();
ensureFolderStructureExists(Path.Combine(Path.Combine(tmppath, speakerName), "speech.mp3"));
string file = Path.Combine(Path.Combine(tmppath, speakerName), "speech" + hash + ".mp3");
FileInfo fileInfo = new FileInfo(file);
if (!fileInfo.Exists)
{
SynthesizeSpeechRequest sreq = new SynthesizeSpeechRequest();
sreq.Text = currentText;
sreq.OutputFormat = OutputFormat.Mp3;
sreq.VoiceId = currentVoice;
SynthesizeSpeechResponse sres = pc.SynthesizeSpeech(sreq);
using (var fileStream = File.Create(file))
{
sres.AudioStream.CopyTo(fileStream);
fileStream.Flush();
fileStream.Close();
}
}
MediaPlayer.Stop();
if (Game1.activeClickableMenu is DialogueBox || Game1.hudMessages.Count > 0 || speak)
{
speak = false;
MediaPlayer.Play(Song.FromUri("speech" + hash, new System.Uri(Path.Combine(Path.Combine(Path.Combine("Content", "TTS"), speakerName), "speech" + hash + ".mp3"), System.UriKind.Relative)));
}
lastText = currentText;
}
}
/// <summary>Snippet for SynthesizeSpeech</summary>
public void SynthesizeSpeech()
{
// Snippet: SynthesizeSpeech(SynthesisInput, VoiceSelectionParams, AudioConfig, CallSettings)
// Create client
TextToSpeechClient textToSpeechClient = TextToSpeechClient.Create();
// Initialize request argument(s)
SynthesisInput input = new SynthesisInput();
VoiceSelectionParams voice = new VoiceSelectionParams();
AudioConfig audioConfig = new AudioConfig();
// Make the request
SynthesizeSpeechResponse response = textToSpeechClient.SynthesizeSpeech(input, voice, audioConfig);
// End snippet
}
public void Start(RichTextBox richText, string AccessKeyID, string SecretAccessKey, VoiceId voiceId)
{
BasicAWSCredentials awsCredentials =
new BasicAWSCredentials(AccessKeyID, SecretAccessKey);
//"AccessKeyID", "SecretAccessKey"
AmazonPollyClient amazonPollyClient =
new AmazonPollyClient(awsCredentials, RegionEndpoint.EUCentral1);
SynthesizeSpeechRequest synthesizeSpeechRequest = MakeSynthesizeSpeechRequest(richText.Text, voiceId);
SynthesizeSpeechResponse synthesizeSpeechResponse =
amazonPollyClient.SynthesizeSpeech(synthesizeSpeechRequest);
CreateMp3File(synthesizeSpeechResponse.AudioStream);
}
private void button1_Click(object sender, EventArgs e)
{
string inpTemp = input.Text;
string inpStr = xuly(inpTemp);
long inp;
bool success = Int64.TryParse(inpStr, out inp);
if (viBtn.Checked == true)
{
String payload = ChuyenSoSangChuoi(inp);
TextToSpeechClient client = TextToSpeechClient.Create();
SynthesizeSpeechResponse response = client.SynthesizeSpeech(
new SynthesisInput()
{
Text = payload
},
new VoiceSelectionParams()
{
LanguageCode = "vi-VN",
Name = "vi-VN-Standard-A"
},
new AudioConfig()
{
AudioEncoding = AudioEncoding.Linear16
}
);
string speechFile = Path.Combine(Directory.GetCurrentDirectory(), "sample.wav");
File.WriteAllBytes(speechFile, response.AudioContent.ToByteArray());
System.Media.SoundPlayer player = new System.Media.SoundPlayer(speechFile);
player.Play();
}
else
{
var synthesizer = new SpeechSynthesizer();
synthesizer.SetOutputToDefaultAudioDevice();
synthesizer.Speak(NumberToWords(inp));
}
}
/// <summary>Snippet for SynthesizeSpeech</summary>
public void SynthesizeSpeech_RequestObject()
{
// Snippet: SynthesizeSpeech(SynthesizeSpeechRequest, CallSettings)
// Create client
TextToSpeechClient textToSpeechClient = TextToSpeechClient.Create();
// Initialize request argument(s)
SynthesizeSpeechRequest request = new SynthesizeSpeechRequest
{
Input = new SynthesisInput(),
Voice = new VoiceSelectionParams(),
AudioConfig = new AudioConfig(),
};
// Make the request
SynthesizeSpeechResponse response = textToSpeechClient.SynthesizeSpeech(request);
// End snippet
}
/// <summary>
/// Unmarshaller the response from the service to the response class.
/// </summary>
/// <param name="context"></param>
/// <returns></returns>
public override AmazonWebServiceResponse Unmarshall(JsonUnmarshallerContext context)
{
SynthesizeSpeechResponse response = new SynthesizeSpeechResponse();
response.AudioStream = context.Stream;
if (context.ResponseData.IsHeaderPresent("Content-Type"))
{
response.ContentType = context.ResponseData.GetHeaderValue("Content-Type");
}
if (context.ResponseData.IsHeaderPresent("x-amzn-RequestCharacters"))
{
response.RequestCharacters = int.Parse(context.ResponseData.GetHeaderValue("x-amzn-RequestCharacters"), CultureInfo.InvariantCulture);
}
return(response);
}
/// <summary>Snippet for SynthesizeSpeechAsync</summary>
public async Task SynthesizeSpeechAsync()
{
// Snippet: SynthesizeSpeechAsync(SynthesisInput, VoiceSelectionParams, AudioConfig, CallSettings)
// Additional: SynthesizeSpeechAsync(SynthesisInput, VoiceSelectionParams, AudioConfig, CancellationToken)
// Create client
TextToSpeechClient textToSpeechClient = await TextToSpeechClient.CreateAsync();
// Initialize request argument(s)
SynthesisInput input = new SynthesisInput();
VoiceSelectionParams voice = new VoiceSelectionParams();
AudioConfig audioConfig = new AudioConfig();
// Make the request
SynthesizeSpeechResponse response = await textToSpeechClient.SynthesizeSpeechAsync(input, voice, audioConfig);
// End snippet
}
public static void runPolly(string inputText) //TODO make polly config settings accessible from editor
{
SynthesizeSpeechRequest sreq = new SynthesizeSpeechRequest();
sreq.Text = inputText;
sreq.OutputFormat = OutputFormat.Mp3;
sreq.VoiceId = VoiceId.Matthew;
SynthesizeSpeechResponse sres = pc.SynthesizeSpeech(sreq);
using (var fileStream = File.Create(@"Output.mp3"))
{
sres.AudioStream.CopyTo(fileStream);
fileStream.Flush();
fileStream.Close();
}
}
public async static Task <string> PollyDemo(IAmazonPolly pollyClient, IAmazonS3 S3Client, string text)
{
string result = null;
SynthesizeSpeechRequest synthesizeRequest = new SynthesizeSpeechRequest()
{
LanguageCode = LanguageCode.EnUS,
OutputFormat = "mp3",
SampleRate = "8000",
Text = text,
TextType = "text",
VoiceId = "Joanna"
};
try
{
Task <SynthesizeSpeechResponse> synthesizeTask = pollyClient.SynthesizeSpeechAsync(synthesizeRequest);
SynthesizeSpeechResponse syntheizeResponse = await synthesizeTask;
Console.WriteLine(syntheizeResponse.ContentType);
Console.WriteLine(syntheizeResponse.RequestCharacters);
using (MemoryStream ms = new MemoryStream())
{
syntheizeResponse.AudioStream.CopyTo(ms);
Console.WriteLine(ms.Length);
// Upload image to S3 bucket
string bucketName = "reinvent-indiamazones";
//string key = dto.text;
string key = "pollytest";
await Task.Run(() => S3Util.UploadToS3(S3Client, bucketName, key, ms));
// TODO : need to check the file exists in S3
result = S3Util.GetPresignedURL(S3Client, bucketName, key);
}
//syntheizeResponse.AudioStream.CopyTo(result);
//result.Flush();
}
catch (AmazonPollyException pollyException)
{
Console.WriteLine(pollyException.Message, pollyException.InnerException);
}
return(result);
}
public static string Speak(string txt, QuestToSpeech.Voice voice, string filePath, AwsAPIConfig config)
{
try {
using (AmazonPollyClient client = new AmazonPollyClient(config.AccessKey, config.SecretKey, Amazon.RegionEndpoint.GetBySystemName(config.Region))) {
string nameWoLang = voice.Name.Replace(voice.LangCode + "-", "");
SynthesizeSpeechRequest req = new SynthesizeSpeechRequest()
{
OutputFormat = OutputFormat.Mp3,
LanguageCode = voice.LangCode,
VoiceId = (VoiceId)typeof(VoiceId).GetField(nameWoLang).GetValue(null),
Text = txt
};
using (FileStream fileStream = new FileStream(filePath + ".mp3", FileMode.Create, FileAccess.Write)) {
SynthesizeSpeechResponse res = client.SynthesizeSpeech(req);
byte[] buffer = new byte[2 * 1024];
int readBytes;
using (Stream audioStream = res.AudioStream) {
while ((readBytes = audioStream.Read(buffer, 0, 2 * 1024)) > 0)
{
fileStream.Write(buffer, 0, readBytes);
}
}
}
// convert to wav and delete mp3
if (File.Exists(filePath + ".mp3"))
{
using (Mp3FileReader mp3 = new Mp3FileReader(filePath + ".mp3")) {
using (WaveStream pcm = WaveFormatConversionStream.CreatePcmStream(mp3)) {
WaveFileWriter.CreateWaveFile(filePath, pcm);
}
}
File.Delete(filePath + ".mp3");
}
}
} catch (Exception ex) {
return(ex.InnerException == null ? ex.Message : ex.InnerException.ToString());
}
return(null);
}
public static void TestPolly()
{
using (AmazonPollyClient pc = new AmazonPollyClient())
{
SynthesizeSpeechRequest sreq = new SynthesizeSpeechRequest();
sreq.Text = "Your Sample Text Here 123";
sreq.OutputFormat = OutputFormat.Mp3;
sreq.VoiceId = VoiceId.Salli;
SynthesizeSpeechResponse sres = pc.SynthesizeSpeechAsync(sreq).GetAwaiter().GetResult();
using (var fileStream = File.Create(@".\TestAudio\output-from-polly.mp3"))
{
sres.AudioStream.CopyTo(fileStream);
fileStream.Flush();
fileStream.Close();
}
}
}
private void Speak(string text)
{
if (client == null)
{
return;
}
// Set the text input to be synthesized.
SynthesisInput input = new SynthesisInput
{
Text = text
};
// Select the type of audio file you want returned.
AudioConfig config = new AudioConfig
{
AudioEncoding = AudioEncoding.Mp3,
// Pitch = -5,
SpeakingRate = Rate
};
// Perform the Text-to-Speech request, passing the text input
// with the selected voice parameters and audio file type
SynthesizeSpeechResponse response = client.SynthesizeSpeech(new SynthesizeSpeechRequest
{
Input = input,
Voice = new VoiceSelectionParams
{
LanguageCode = languages[Settings.Default.languageIndex],
Name = voiceTypes[Settings.Default.useWavenetVoices ? 0 : 1, Settings.Default.languageIndex, Settings.Default.ttsVoice]
},
AudioConfig = config
});
// Write the AudioContent of the response to an MP3 file.
string s64 = response.AudioContent.ToBase64();
string filePath = Path.Combine(Path.GetTempPath(), "ttsoutput64_" + DateTime.Now.ToFileTime() + ".mp3");
File.WriteAllBytes(filePath, Convert.FromBase64String(s64));
// play the file
ttsQueue.Enqueue(filePath);
}
/// <summary>
/// Returns temporary file with MP3 for input text, synthetized in call to Amazon Polly.
/// </summary>
/// <param name="text">Text to become speech.</param>
/// <param name="voice">Voice. Default is Joanna in English, US.</param>
/// <returns></returns>
static public string GetSpeech(string text, string voice = "")
{
SynthesizeSpeechRequest speechRequest = new SynthesizeSpeechRequest();
speechRequest.Text = text;
speechRequest.OutputFormat = OutputFormat.Mp3;
speechRequest.VoiceId = VoiceIdFromString(voice);
SynthesizeSpeechResponse speech = PollyClient.SynthesizeSpeech(speechRequest);
string speechFilename = Path.Combine(Path.GetTempPath(), DateTime.Now.Ticks.ToString(CultureInfo.InvariantCulture) + "_" + Guid.NewGuid().ToString() + ".mp3");
using (var fileStream = File.Create(speechFilename))
{
speech.AudioStream.CopyTo(fileStream);
fileStream.Flush();
}
return(speechFilename);
}
/// <summary>
/// Google CloudからのレスポンスをAudioClipに書き出し，再生する
/// </summary>
/// <param name="response">Google Cloudからのレスポンス</param>
private static void SetAudioClip(SynthesizeSpeechResponse response)
{
var bytes = response.AudioContent.ToByteArray();
// byte[]をAudioClipで利用できる形に変換する
var wav = new WAV(bytes);
Debug.Log("Get Response: Elapsed time " + _stopwatch.ElapsedMilliseconds + "ms.\nData Length: " +
(wav.SampleCount * (1f / wav.Frequency) * 1000f).ToString("F0") + "ms.");
_context.Post(_ =>
{
// AudioSourceに新しいAudioClipを貼り付ける
audioSource.clip = AudioClip.Create("TextToSpeech", wav.SampleCount, 1, wav.Frequency, false);
audioSource.clip.SetData(wav.LeftChannel, 0);
// AudioClipを再生
audioSource.Play();
}, null);
}
public async Task ProcessMessageAsync(SQSEvent.SQSMessage message)
{
//_logger.LogLine("Processing message \n" + message.Body);
ConvertTextToSpeechContract contract =
JsonConvert.DeserializeObject <ConvertTextToSpeechContract>(message.Body);
DetectDominantLanguageResponse dominantLanguageResponse = AwsService.Comprehend
.DetectDominantLanguage(contract.TextContent);
string dominantLanguageCode = dominantLanguageResponse
.Languages
.OrderByDescending(l => l.Score)
.First()
.LanguageCode;
Amazon.Polly.LanguageCode languageCode = null;
switch (dominantLanguageCode)
{
case "en":
languageCode = Amazon.Polly.LanguageCode.EnUS;
break;
//TODO: Handle other languages
default:
throw new Exception("Could not find the language specified");
}
SynthesizeSpeechResponse systhesisResponse = AwsService
.Polly.SynthesizeSpeech(contract.TextContent, languageCode);
MemoryStream inputStream = new MemoryStream(ReadToEnd(systhesisResponse.AudioStream));
string contentType = "audio/mpeg";
string bucketName = Environment.GetEnvironmentVariable("ALEXA_READER_BUCKET");
AwsService.S3.PutObject(inputStream, contract.AudioFilePathToSave,
bucketName, contentType);
//_logger.LogLine("Processed message \n" + message.Body);
}
private static async Task RecordFile()
{
AmazonPollyClient pc = new AmazonPollyClient(AmazonCredHelper.AwsCredintals, Amazon.RegionEndpoint.USEast2);
SynthesizeSpeechRequest sreq = new SynthesizeSpeechRequest();
sreq.Text = textToVoice;
sreq.OutputFormat = OutputFormat.Mp3;
sreq.VoiceId = VoiceId.Joey;
SynthesizeSpeechResponse sres = await pc.SynthesizeSpeechAsync(sreq);
var path = Path.Combine(dirpath, outputFilename + ".mp3");
using (var fileStream = File.Create(path))
{
sres.AudioStream.CopyTo(fileStream);
fileStream.Flush();
fileStream.Close();
}
}
public static void Main(string[] args)
{
AmazonPollyClient client = new AmazonPollyClient("yourID", "yourSecretKey", RegionEndpoint.USEast1);
SynthesizeSpeechRequest request = new SynthesizeSpeechRequest();
request.OutputFormat = OutputFormat.Mp3;
request.Text = "This is my first conversion";
request.TextType = TextType.Text;
request.VoiceId = VoiceId.Nicole;
SynthesizeSpeechResponse response = client.SynthesizeSpeech(request);
Console.WriteLine("ContentType: " + response.ContentType);
Console.WriteLine("RequestCharacters: " + response.RequestCharacters);
FileStream destination = File.Open(@"c:\temp\myfirstconversion.mp3", FileMode.Create);
response.AudioStream.CopyTo(destination);
Console.WriteLine("Destination length: {0}", destination.Length.ToString());
destination.Close();
Console.Read();
}
