Text-To-Speech for .NET Micro Framework

Posted by Marco Minerva

I’m very interested in the possibility of making my applications speak. So, I have realized a library for .NET Micro Framework that uses Microsoft Translator to obtain streams of wave-file speaking text in the desired language. At this moment, there are 44 supported languages, including English, Italian, German, French, Spanish, Japanese and Chinese.

To run the code, you need to get a Bing Application ID, that you can get from the page http://msdn.microsoft.com/en-us/library/ff512386.aspx. Library usage is very simple:

SpeechSynthesizer speech = new SpeechSynthesizer(APP_ID);

string text = "Have a nice day!";
string language = "en";

bytes[] data = speech.GetSpeakBytes(text, language);

In this way, we obtain a byte array that contains the stream of a wave-file speaking the passed-in text in the desired language. We can then reproduce it using our device audio functionalities, for example the .NET Gadgeteer Music Module from GHI Electronics.

The above method is blocking, but there is also a corresponding asynchronous version, GetSpeakBytesAsync.

SpeechSynthesizer speech = new SpeechSynthesizer(APP_ID);
speech.GetSpeakBytesCompleted +=
                new SpeechSynthesizer.GetSpeakBytesEventHandler(speech_GetSpeakBytesCompleted);

string text = "Have a nice day!";
string language = "en";

speech.GetSpeakBytesAsync(text, language);    // This method returns immediately

private void speech_GetSpeakBytesCompleted(object sender, GetSpeakBytesEventArgs e)
{
    if (e.Error == null)
        Debug.Print("Retrived a speech of  " + e.Data.Length + " bytes length");
    else
        Debug.Print("There was an error: " + e.Error.Message);
}

This library is available on MSDN Code Gallery.  To guarantee a better diffusion, I have published it on NuGet too (search for TranslatorService.Speech).

TranslatorService.Speech

The library available on NuGet

Finally, note that there is also a version of this library that runs on .NET “Full Framework”. Both versions are available on NuGet, so based on the type of our projects, Visual Studio will reference the correct one.

The code that runs the example in the video is shown in the following block.

using Microsoft.SPOT;

using GT = Gadgeteer;
using GTM = Gadgeteer.Modules;
using Gadgeteer.Modules.GHIElectronics;
using MicroTranslatorService.Speech;
namespace TextToSpeech
{
    public partial class Program
    {
        private const string APP_ID = "Add your Bing user ID";
        SpeechSynthesizer speech;
        byte[] bytes;
        uint index;
        void ProgramStarted()
        {
            button.ButtonPressed += new Button.ButtonEventHandler(button_ButtonPressed);
            music.MusicFinished += new MusicFinishedPlayingEventHandler(music_MusicFinished);
            index = 0;

            speech = new SpeechSynthesizer(APP_ID);
            speech.GetSpeakBytesCompleted += new SpeechSynthesizer.GetSpeakBytesEventHandler(speech_GetSpeakBytesCompleted);

            Debug.Print("Program Started");
        }

        void music_MusicFinished()
        {
            Debug.Print("Audio finished.");
        }

        void button1_ButtonPressed(Button sender, Button.ButtonState state)
        {
            if (bytes != null)
                music.Play(bytes);
        }

        void speech_GetSpeakBytesCompleted(object sender, GetSpeakBytesEventArgs e)
        {
            if (e.Error == null)
            {
                Debug.Print("Retrived a speech of " + e.Data.Length + " bytes length");
                bytes = e.Data;
                music.Play(bytes);
            }
            else
                Debug.Print("There was an error: " + e.Error.Message);
        }

        void SendRequest(string text, string language)
        {
            Debug.Print("Submitting request...");
            display.SimpleGraphics.Clear();
            display.SimpleGraphics.DisplayTextInRectangle(text, 20, 20, 300, 250, GT.Color.Cyan,
                Resources.GetFont(Resources.FontResources.NinaB),
                GTM.Module.DisplayModule.SimpleGraphicsInterface.TextAlign.Left,
                GTM.Module.DisplayModule.SimpleGraphicsInterface.WordWrap.Wrap,
                GTM.Module.DisplayModule.SimpleGraphicsInterface.Trimming.WordEllipsis,
                GTM.Module.DisplayModule.SimpleGraphicsInterface.ScaleText.None);

            speech.GetSpeakBytesAsync(text, language);
        }

        void button_ButtonPressed(Button sender, Button.ButtonState state)
        {
            switch (index)
            {
                case 0:
                    {
                        SendRequest("I am very interested in the possibility of making my applications speak. So, I have realized a library for .NET Micro Framework that uses Microsoft Translator to obtain streams of wave-file speaking text in the desired language. At this moment, there are 44 supported languages, including English, Italian, German, French, Spanish, Japanese and Chinese.", "en");
                        index++;
                        break;
                    }
                case 1:
                    {
                        SendRequest("Je suis très intéressé par la possibilité de rendre mes applications parler. Alors, j'ai réalisé une bibliothèque pour. NET Micro Framework qui utilise Microsoft Translator pour obtenir des flux de texte parlant fichier wave dans la langue désirée. En ce moment, il ya 44 langues prises en charge, y compris l'anglais, italien, allemand, français, espagnol, japonais et chinois.", "fr");
                        index++;
                        break;
                    }

                case 2:
                    {
                        SendRequest("Ich bin sehr in die Möglichkeit, meine Anwendungen sprechen interessiert. Also, ich habe eine Bibliothek für. NET Micro Framework, Microsoft Translator nutzt, um Ströme von wave-Datei sprechenden Text in die gewünschte Sprache zu erhalten realisiert. In diesem Moment gibt es 44 unterstützten Sprachen, darunter Englisch, Italienisch, Deutsch, Französisch, Spanisch, Japanisch und Chinesisch.", "de");
                        index++;
                        break;
                    }
                case 3:
                    {
                        SendRequest("Does this train go to Barcelona?", "es");
                        index++;
                        break;
                    }
                case 4:
                    {
                        SendRequest("Please pass the chocolate.", "pt");
                        index++;
                        break;
                    }
                case 5:
                    {
                        SendRequest("Thanks for all the fish.", "no");
                        index++;
                        break;
                    }
                case 6:
                    {
                        SendRequest("Will the winter never end?", "ru");
                        index++;
                        break;
                    }
                case 7:
                    {
                        //SendRequest("私は非常に私のアプリケーションは、話すことの可能性に興味を持っています。だから、私は希望の言語でウェーブファイルを話すのテキストのストリームを取得するには、Microsoft Translatorを使用している。NET Micro Frameworkのためのライブラリを実現しています。現時点では、英語、イタリア語、ドイツ語、フランス語、スペイン語、日本語、中国語など44言語がサポートさ、があります", "ja");
                        index++;
                        break;
                    }
                default:
                    {
                        index = 0;
                        break;
                    }
            }
        }
    }
}
Advertisement

,

  1. New version of Text-To-Speech library « Integral Design
  2. Controllare il Text-To-Speech da un’applicazione client « Integral Design
  3. Controlling Text-To-Speech from a client application « Integral Design

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: