Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Create shared classlib #10

Merged
merged 10 commits into from
Jan 14, 2024
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension


Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
19 changes: 19 additions & 0 deletions speech-translator.sln
Original file line number Diff line number Diff line change
Expand Up @@ -16,6 +16,12 @@ Project("{2150E333-8FDC-42A3-9474-1A3956D46DE8}") = "ソリューション項目
README.md = README.md
EndProjectSection
EndProject
Project("{9A19103F-16F7-4668-BE54-9A1E7A4F7556}") = "SpeechTranslatorShared", "src\Shared\SpeechTranslatorShared.csproj", "{FF0262A4-994A-4237-88E4-1049E03E85D3}"
EndProject
Project("{2150E333-8FDC-42A3-9474-1A3956D46DE8}") = "tests", "tests", "{00192D0A-A300-4EA4-A6BA-D193D73004D5}"
EndProject
Project("{9A19103F-16F7-4668-BE54-9A1E7A4F7556}") = "SpeechTranslatorShared.Tests", "tests\Shared.Tests\SpeechTranslatorShared.Tests.csproj", "{EA08923B-519F-40CB-B219-4826FE0954F2}"
EndProject
Global
GlobalSection(SolutionConfigurationPlatforms) = preSolution
Debug|Any CPU = Debug|Any CPU
Expand All @@ -26,11 +32,24 @@ Global
{CFD1A73A-09D2-46F2-8E34-0247E0D19A5B}.Debug|Any CPU.Build.0 = Debug|Any CPU
{CFD1A73A-09D2-46F2-8E34-0247E0D19A5B}.Release|Any CPU.ActiveCfg = Release|Any CPU
{CFD1A73A-09D2-46F2-8E34-0247E0D19A5B}.Release|Any CPU.Build.0 = Release|Any CPU
{FF0262A4-994A-4237-88E4-1049E03E85D3}.Debug|Any CPU.ActiveCfg = Debug|Any CPU
{FF0262A4-994A-4237-88E4-1049E03E85D3}.Debug|Any CPU.Build.0 = Debug|Any CPU
{FF0262A4-994A-4237-88E4-1049E03E85D3}.Release|Any CPU.ActiveCfg = Release|Any CPU
{FF0262A4-994A-4237-88E4-1049E03E85D3}.Release|Any CPU.Build.0 = Release|Any CPU
{EA08923B-519F-40CB-B219-4826FE0954F2}.Debug|Any CPU.ActiveCfg = Debug|Any CPU
{EA08923B-519F-40CB-B219-4826FE0954F2}.Debug|Any CPU.Build.0 = Debug|Any CPU
{EA08923B-519F-40CB-B219-4826FE0954F2}.Release|Any CPU.ActiveCfg = Release|Any CPU
{EA08923B-519F-40CB-B219-4826FE0954F2}.Release|Any CPU.Build.0 = Release|Any CPU
EndGlobalSection
GlobalSection(SolutionProperties) = preSolution
HideSolutionNode = FALSE
EndGlobalSection
GlobalSection(NestedProjects) = preSolution
{CFD1A73A-09D2-46F2-8E34-0247E0D19A5B} = {358DED3F-4540-44E2-993E-AA7BB7E61EE6}
{FF0262A4-994A-4237-88E4-1049E03E85D3} = {358DED3F-4540-44E2-993E-AA7BB7E61EE6}
{EA08923B-519F-40CB-B219-4826FE0954F2} = {00192D0A-A300-4EA4-A6BA-D193D73004D5}
EndGlobalSection
GlobalSection(ExtensibilityGlobals) = postSolution
SolutionGuid = {2C0686D0-47ED-4A5B-A104-962D1DA98B29}
EndGlobalSection
EndGlobal
3 changes: 3 additions & 0 deletions src/Shared/GlobalUsings.cs
Original file line number Diff line number Diff line change
@@ -0,0 +1,3 @@
global using Microsoft.CognitiveServices.Speech;
global using Microsoft.CognitiveServices.Speech.Audio;
global using Microsoft.CognitiveServices.Speech.Translation;
13 changes: 13 additions & 0 deletions src/Shared/SpeechTranslatorShared.csproj
Original file line number Diff line number Diff line change
@@ -0,0 +1,13 @@
<Project Sdk="Microsoft.NET.Sdk">

<PropertyGroup>
<TargetFramework>net8.0</TargetFramework>
<ImplicitUsings>enable</ImplicitUsings>
<Nullable>enable</Nullable>
</PropertyGroup>

<ItemGroup>
<PackageReference Include="Microsoft.CognitiveServices.Speech" Version="1.33.0" />
</ItemGroup>

</Project>
18 changes: 18 additions & 0 deletions src/Shared/TranslationRecognizerWorkerBase.cs
Original file line number Diff line number Diff line change
@@ -0,0 +1,18 @@
namespace SpeechTranslatorShared;

public abstract class TranslationRecognizerWorkerBase
{
public abstract void OnRecognizing(TranslationRecognitionEventArgs e);

public abstract void OnRecognized(TranslationRecognitionEventArgs e);

public abstract void OnCanceled(TranslationRecognitionCanceledEventArgs e);

public abstract void OnSpeechStartDetected(RecognitionEventArgs e);

public abstract void OnSpeechEndDetected(RecognitionEventArgs e);

public abstract void OnSessionStarted(SessionEventArgs e);

public abstract void OnSessionStopped(SessionEventArgs e);
}
77 changes: 77 additions & 0 deletions src/Shared/Translator.cs
Original file line number Diff line number Diff line change
@@ -0,0 +1,77 @@
namespace SpeechTranslatorShared;

public class Translator
{
private readonly SpeechTranslationConfig _speechTranslationConfig;

public Translator(Uri endpointUrl, string subscriptionKey, string recognitionLanguage = "en-US", string targetLanguage = "ja-JP")
{
if (endpointUrl is null)
{
throw new ArgumentNullException(nameof(endpointUrl));
}

if (string.IsNullOrWhiteSpace(subscriptionKey))
{
throw new ArgumentException($"'{nameof(subscriptionKey)}' を NULL または空にすることはできません。", nameof(subscriptionKey));
}

if (string.IsNullOrWhiteSpace(recognitionLanguage))
{
throw new ArgumentException($"'{nameof(recognitionLanguage)}' を NULL または空にすることはできません。", nameof(recognitionLanguage));
}

if (string.IsNullOrWhiteSpace(targetLanguage))
{
throw new ArgumentException($"'{nameof(targetLanguage)}' を NULL または空にすることはできません。", nameof(targetLanguage));
}

_speechTranslationConfig = SpeechTranslationConfig.FromEndpoint(endpointUrl, subscriptionKey);
_speechTranslationConfig.SpeechRecognitionLanguage = recognitionLanguage;
_speechTranslationConfig.AddTargetLanguage(targetLanguage);
_speechTranslationConfig.SetProperty(PropertyId.SpeechServiceConnection_TranslationVoice, "de-DE-Hedda");
}

public async Task MultiLingualTranslation(TranslationRecognizerWorkerBase worker)
{
if (worker is null)
{
throw new ArgumentNullException(nameof(worker));
}

var autoDetectSourceLanguageConfig = AutoDetectSourceLanguageConfig.FromLanguages([_speechTranslationConfig.SpeechRecognitionLanguage]);
var stopTranslation = new TaskCompletionSource<int>();

using (var audioInput = AudioConfig.FromDefaultMicrophoneInput())
using (var recognizer = new TranslationRecognizer(_speechTranslationConfig, autoDetectSourceLanguageConfig, audioInput))
{
recognizer.Recognizing += (s, e) => worker.OnRecognizing(e);

recognizer.Recognized += (s, e) => worker.OnRecognized(e);

recognizer.Canceled += (s, e) =>
{
stopTranslation.TrySetResult(0);
worker.OnCanceled(e);
};

recognizer.SpeechStartDetected += (s, e) => worker.OnSpeechStartDetected(e);

recognizer.SpeechEndDetected += (s, e) => worker.OnSpeechEndDetected(e);

recognizer.SessionStarted += (s, e) => worker.OnSessionStarted(e);

recognizer.SessionStopped += (s, e) =>
{
stopTranslation.TrySetResult(0);
worker.OnSessionStopped(e);
};

// Starts continuous recognition. Uses StopContinuousRecognitionAsync() to stop recognition.
await recognizer.StartContinuousRecognitionAsync().ConfigureAwait(false);

Task.WaitAny(new[] { stopTranslation.Task });
await recognizer.StopContinuousRecognitionAsync().ConfigureAwait(false);
}
}
}
9 changes: 9 additions & 0 deletions src/SpeechTranslatorConsole/GlobalUsings.cs
Original file line number Diff line number Diff line change
@@ -0,0 +1,9 @@
global using System.Text;

global using Microsoft.CognitiveServices.Speech;
global using Microsoft.CognitiveServices.Speech.Translation;
global using Microsoft.Extensions.Configuration;

global using SpeechTranslatorConsole;

global using SpeechTranslatorShared;
133 changes: 6 additions & 127 deletions src/SpeechTranslatorConsole/Program.cs
Original file line number Diff line number Diff line change
@@ -1,12 +1,4 @@
using System.Text;
using Microsoft.CognitiveServices.Speech;
using Microsoft.CognitiveServices.Speech.Audio;
using Microsoft.CognitiveServices.Speech.Translation;
using Microsoft.Extensions.Configuration;

using SpeechTranslatorConsole;

const string directoryName = "recordings";
const string directoryName = "recordings";

var environmentName = Environment.GetEnvironmentVariable("DOTNETCORE_ENVIRONMENT") ?? "Development";

Expand All @@ -18,6 +10,7 @@

Settings? settings = config.GetRequiredSection(nameof(Settings)).Get<Settings>();
var region = settings?.Region ?? throw new ArgumentNullException("region");
var endpointUrl = new Uri($"wss://{region}.stt.speech.microsoft.com/speech/universal/v2");
var subscriptionKey = settings?.SubscriptionKey ?? throw new ArgumentNullException("subscriptionKey");
var fromLanguage = DetermineLanguage("speaker") ?? throw new ArgumentNullException("fromLanguage");
Console.WriteLine();
Expand All @@ -26,23 +19,20 @@

Console.Write("Record file name: ");
var filePath = Console.ReadLine() ?? throw new ArgumentNullException("filePath");
Console.WriteLine();

if (!Directory.Exists(directoryName))
{
Directory.CreateDirectory(directoryName);
Console.WriteLine("The directory was created successfully at {0}.", Directory.GetCreationTime(directoryName));
}

var endpointString = $"wss://{region}.stt.speech.microsoft.com/speech/universal/v2";
var endpointUrl = new Uri(endpointString);
var speechTranslationConfig = SpeechTranslationConfig.FromEndpoint(endpointUrl, subscriptionKey);
speechTranslationConfig.SpeechRecognitionLanguage = fromLanguage;
speechTranslationConfig.AddTargetLanguage(targetLanguage);
speechTranslationConfig.SetProperty(PropertyId.SpeechServiceConnection_TranslationVoice, "de-DE-Hedda");
var translator = new Translator(endpointUrl, subscriptionKey, fromLanguage, targetLanguage);
var worker = new TranslationRecognizerWorker($"{directoryName}/{filePath}.txt");

try
{
await MultiLingualTranslation($"{directoryName}/{filePath}.txt", speechTranslationConfig);
await translator.MultiLingualTranslation(worker);
}
catch (Exception e)
{
Expand All @@ -66,114 +56,3 @@
_ => null
};
}

static async Task MultiLingualTranslation(string filepath, SpeechTranslationConfig config)
{
var autoDetectSourceLanguageConfig = AutoDetectSourceLanguageConfig.FromLanguages(new string[] { config.SpeechRecognitionLanguage });

var stopTranslation = new TaskCompletionSource<int>();
using (var audioInput = AudioConfig.FromDefaultMicrophoneInput())
{
using (var recognizer = new TranslationRecognizer(config, autoDetectSourceLanguageConfig, audioInput))
{
recognizer.Recognizing += (s, e) =>
{
Console.Write(".");

//var lidResult = e.Result.Properties.GetProperty(PropertyId.SpeechServiceConnection_AutoDetectSourceLanguageResult);

//Console.WriteLine($"RECOGNIZING in '{lidResult}': Text={e.Result.Text}");
//foreach (var element in e.Result.Translations)
//{
// Console.WriteLine($" TRANSLATING into '{element.Key}': {element.Value}");
//}
};

recognizer.Recognized += (s, e) =>
{
Console.WriteLine();
Console.WriteLine();

if (e.Result.Reason == ResultReason.TranslatedSpeech)
{
var sw = new StreamWriter(filepath, true, Encoding.UTF8);

var lidResult = e.Result.Properties.GetProperty(PropertyId.SpeechServiceConnection_AutoDetectSourceLanguageResult);

// Console.WriteLine($"RECOGNIZED in '{lidResult}': Text={e.Result.Text}");
Console.WriteLine($"{e.Result.Text}");
sw.WriteLine($"{e.Result.Text}");
foreach (var element in e.Result.Translations)
{
// Console.WriteLine($" TRANSLATED into '{element.Key}': {element.Value}");
Console.WriteLine($"{element.Value}");
sw.WriteLine($"{element.Value}");
}

sw.WriteLine();
sw.Close();
}
else if (e.Result.Reason == ResultReason.RecognizedSpeech)
{
Console.WriteLine($"RECOGNIZED: Text={e.Result.Text}");
Console.WriteLine($" Speech not translated.");
}
else if (e.Result.Reason == ResultReason.NoMatch)
{
Console.WriteLine($"NOMATCH: Speech could not be recognized.");
}

Console.WriteLine();
};

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 set the speech resource key and region values?");
}

stopTranslation.TrySetResult(0);
Console.WriteLine();
};

recognizer.SpeechStartDetected += (s, e) =>
{
Console.WriteLine("\nSpeech start detected event.");
Console.WriteLine();
};

recognizer.SpeechEndDetected += (s, e) =>
{
Console.WriteLine("\nSpeech end detected event.");
Console.WriteLine();
};

recognizer.SessionStarted += (s, e) =>
{
Console.WriteLine("\nSession started event.");
Console.WriteLine();
};

recognizer.SessionStopped += (s, e) =>
{
Console.WriteLine("\nSession stopped event.");
Console.WriteLine($"\nStop translation.");
Console.WriteLine();
stopTranslation.TrySetResult(0);
};

// Starts continuous recognition. Uses StopContinuousRecognitionAsync() to stop recognition.
Console.WriteLine("Start translation...");
Console.WriteLine();
await recognizer.StartContinuousRecognitionAsync().ConfigureAwait(false);

Task.WaitAny(new[] { stopTranslation.Task });
await recognizer.StopContinuousRecognitionAsync().ConfigureAwait(false);
}
}
}
4 changes: 4 additions & 0 deletions src/SpeechTranslatorConsole/SpeechTranslatorConsole.csproj
Original file line number Diff line number Diff line change
Expand Up @@ -15,6 +15,10 @@
<PackageReference Include="Microsoft.Extensions.Configuration.EnvironmentVariables" Version="8.0.0" />
</ItemGroup>

<ItemGroup>
<ProjectReference Include="..\Shared\SpeechTranslatorShared.csproj" />
</ItemGroup>

<ItemGroup>
<None Update="appsettings.Development.json">
<CopyToOutputDirectory>Always</CopyToOutputDirectory>
Expand Down
Loading
Loading