重新想象 Windows 8.1 Store Apps (87) - TTS: Speak Text, Speak SSML

浏览数：16 / 时间：2015年06月11日

作者：webabcd

介绍
重新想象 Windows 8.1 Store Apps 之 TTS（Text To Speech）

Speak Text
Speak SSML

示例
1、演示如何通过 TTS 朗读一段文本，以及如何将其保存为音频文件
SpeakText.xaml

<Page
    x:Class="Windows81.TTS.SpeakText"
    xmlns="http://schemas.microsoft.com/winfx/2006/xaml/presentation"
    xmlns:x="http://schemas.microsoft.com/winfx/2006/xaml"
    xmlns:local="using:Windows81.TTS"
    xmlns:d="http://schemas.microsoft.com/expression/blend/2008"
    xmlns:mc="http://schemas.openxmlformats.org/markup-compatibility/2006"
    mc:Ignorable="d">

    <Grid Background="Transparent">
        <StackPanel Margin="120 0 0 0">

            <!--用于播放音频-->
            <MediaElement Name="mediaElement" AutoPlay="False" />

            <TextBlock Name="lblMsg" FontSize="14.667" Margin="0 10 0 0" />
            
            <!--需要被 TTS 的文本-->
            <TextBlock Name="lblText" FontSize="14.667" Margin="0 10 0 0" Text="TTS 是 Text To Speech 的缩写，即“从文本到语音”，是人机对话的一部分，让机器能够说话。" />
            
            <!--开始 TTS-->
            <Button Name="btnSpeakText" Content="Speek Text" Click="btnSpeakText_Click" Margin="0 10 0 0" />

            <!--保存 TTS 转换出的音频文件-->
            <Button Name="btnSaveAudio" Content="Save Audio" Click="btnSaveAudio_Click" Margin="0 10 0 0" />

        </StackPanel>
    </Grid>
</Page>

SpeakText.xaml.cs

/*
 * 演示如何通过 TTS 朗读一段文本，以及如何将其保存为音频文件
 * 
 * 注：Windows Phone 中的“TTS, 语音识别, 语音命令”请参见：http://www.cnblogs.com/webabcd/archive/2014/01/02/3501356.html
 */

using System;
using System.Collections.Generic;
using Windows.Media.SpeechSynthesis;
using Windows.Storage;
using Windows.Storage.Pickers;
using Windows.Storage.Streams;
using Windows.UI.Popups;
using Windows.UI.Xaml;
using Windows.UI.Xaml.Controls;
using Windows.UI.Xaml.Media;

namespace Windows81.TTS
{
    public sealed partial class SpeakText : Page
    {
        // TTS 引擎
        private SpeechSynthesizer _synthesizer = new SpeechSynthesizer();

        public SpeakText()
        {
            this.InitializeComponent();

            lblMsg.Text = "系统支持的语音有：";
            // TTS 引擎所支持的全部语音信息
            var voices = SpeechSynthesizer.AllVoices;
            foreach (VoiceInformation voice in voices)
            {
                lblMsg.Text += voice.DisplayName + ", ";
            }

            lblMsg.Text += Environment.NewLine;
            // 获取或设置当前 TTS 引擎所使用的语音
            lblMsg.Text += "当前 TTS 使用的语音是：" +　_synthesizer.Voice.DisplayName;

            // SpeechSynthesizer 实现了 IDisposable 接口
            // _synthesizer.Dispose();

            this.mediaElement.CurrentStateChanged += mediaElement_CurrentStateChanged; 
        }

        void mediaElement_CurrentStateChanged(object sender, RoutedEventArgs e)
        {
            if (mediaElement.CurrentState == MediaElementState.Paused)
            {
                btnSpeakText.IsEnabled = true;
            }
        }

        // 播放 TTS 音频
        private async void btnSpeakText_Click(object sender, RoutedEventArgs e)
        {
            // TTS 引擎生成的音频流
            SpeechSynthesisStream synthesisStream;

            try
            {
                // 通过 TTS 引擎将字符串转换成音频流
                synthesisStream = await _synthesizer.SynthesizeTextToStreamAsync(lblText.Text);
            }
            catch (Exception ex)
            {
                lblMsg.Text = ex.ToString();

                synthesisStream = null;
                btnSpeakText.IsEnabled = true;
            }

            // 播放音频流
            mediaElement.AutoPlay = true;
            mediaElement.SetSource(synthesisStream, synthesisStream.ContentType);
            mediaElement.Play();
        }

        // 保存 TTS 音频
        private async void btnSaveAudio_Click(object sender, RoutedEventArgs e)
        {
            FileSavePicker savePicker = new FileSavePicker();
            savePicker.DefaultFileExtension = ".wav";
            savePicker.FileTypeChoices.Add("Audio file", new List<string>() { ".wav" });

            StorageFile file = await savePicker.PickSaveFileAsync();
            if (file != null)
            {
                btnSaveAudio.IsEnabled = false;

                SpeechSynthesisStream synthesisStream;
                try
                {
                    synthesisStream = await _synthesizer.SynthesizeTextToStreamAsync(lblText.Text);
                }
                catch (Exception ex)
                {
                    lblMsg.Text = ex.ToString();

                    synthesisStream = null;
                    btnSaveAudio.IsEnabled = true;
                }

                // 保存音频数据到文件
                Windows.Storage.Streams.Buffer buffer = new Windows.Storage.Streams.Buffer(4096);
                IRandomAccessStream writeStream = (IRandomAccessStream)await file.OpenAsync(FileAccessMode.ReadWrite);
                IOutputStream outputStream = writeStream.GetOutputStreamAt(0);
                DataWriter dataWriter = new DataWriter(outputStream);
            
                while (synthesisStream.Position < synthesisStream.Size)
                {
                    await synthesisStream.ReadAsync(buffer, 4096, InputStreamOptions.None);
                    dataWriter.WriteBuffer(buffer);
                }

                dataWriter.StoreAsync().AsTask().Wait();
                outputStream.FlushAsync().AsTask().Wait();

                btnSaveAudio.IsEnabled = true;
            }
        }
    }
}

2、演示如何通过 TTS 朗读 SSML 协议文档，以及如何将其保存为音频文件
SpeakSSML.xaml

<Page
    x:Class="Windows81.TTS.SpeakSSML"
    xmlns="http://schemas.microsoft.com/winfx/2006/xaml/presentation"
    xmlns:x="http://schemas.microsoft.com/winfx/2006/xaml"
    xmlns:local="using:Windows81.TTS"
    xmlns:d="http://schemas.microsoft.com/expression/blend/2008"
    xmlns:mc="http://schemas.openxmlformats.org/markup-compatibility/2006"
    mc:Ignorable="d">

    <Grid Background="Transparent">
        <StackPanel Margin="120 0 0 0">

            <!--用于播放音频-->
            <MediaElement Name="mediaElement" AutoPlay="False" />

            <TextBlock Name="lblMsg" FontSize="14.667" Margin="0 10 0 0" />

            <!--开始 TTS-->
            <Button Name="btnSpeakSSML" Content="Speek SSML" Click="btnSpeakSSML_Click" Margin="0 10 0 0" />

            <!--保存 TTS 转换出的音频文件-->
            <Button Name="btnSaveAudio" Content="Save Audio" Click="btnSaveAudio_Click" Margin="0 10 0 0" />

        </StackPanel>
    </Grid>
</Page>

SpeakSSML.xaml.cs

/*
 * 演示如何通过 TTS 朗读 SSML 协议文档，以及如何将其保存为音频文件
 * 
 * 注：Windows Phone 中的“TTS, 语音识别, 语音命令”请参见：http://www.cnblogs.com/webabcd/archive/2014/01/02/3501356.html
 * 
 * 
 * 另外：
 * 1、SSML - Speech Synthesis Markup Language
 * 2、微软关于 ssml 的说明：http://msdn.microsoft.com/en-us/library/hh361578
 * 3、W3C 关于 ssml 的说明：http://www.w3.org/TR/speech-synthesis/
 */

using System;
using System.Collections.Generic;
using Windows.Media.SpeechSynthesis;
using Windows.Storage;
using Windows.Storage.Pickers;
using Windows.Storage.Streams;
using Windows.UI.Popups;
using Windows.UI.Xaml;
using Windows.UI.Xaml.Controls;
using Windows.UI.Xaml.Media;

namespace Windows81.TTS
{
    public sealed partial class SpeakSSML : Page
    {
        // TTS 引擎
        private SpeechSynthesizer _synthesizer = new SpeechSynthesizer();

        public SpeakSSML()
        {
            this.InitializeComponent();
            
            lblMsg.Text = "系统支持的语音有：";
            // TTS 引擎所支持的全部语音信息
            var voices = SpeechSynthesizer.AllVoices;
            foreach (VoiceInformation voice in voices)
            {
                lblMsg.Text += voice.DisplayName + ", ";
            }

            lblMsg.Text += Environment.NewLine;
            // 获取或设置当前 TTS 引擎所使用的语音
            lblMsg.Text += "当前 TTS 使用的语音是：" + _synthesizer.Voice.DisplayName;

            // SpeechSynthesizer 实现了 IDisposable 接口
            // _synthesizer.Dispose();

            this.mediaElement.CurrentStateChanged += mediaElement_CurrentStateChanged;
        }

        void mediaElement_CurrentStateChanged(object sender, RoutedEventArgs e)
        {
            if (mediaElement.CurrentState == MediaElementState.Paused)
            {
                btnSpeakSSML.IsEnabled = true;
            }
        }

        // 播放 TTS 音频（SSML）
        private async void btnSpeakSSML_Click(object sender, RoutedEventArgs e)
        {
            // TTS 引擎生成的音频流
            SpeechSynthesisStream synthesisStream;

            try
            {
                // 构造一个 SSML 协议文档
                string ssml = "<speak version=\"1.0\" xmlns=\"http://www.w3.org/2001/10/synthesis\" xml:lang=\"zh-CN\">"; // 中文
                ssml += "<voice gender=\"male\">"; // 男声
                ssml += "<prosody rate=\"-50%\">"; // 语速放慢 50%
                ssml += "TTS 是 Text To Speech 的缩写，即“从文本到语音”，是人机对话的一部分，让机器能够说话。";
                ssml += "</prosody>";
                ssml += "</voice>";
                ssml += "</speak>";

                // 通过 TTS 引擎将 SSML 协议文档转换成音频流
                synthesisStream = await _synthesizer.SynthesizeSsmlToStreamAsync(ssml);
            }
            catch (Exception ex)
            {
                lblMsg.Text = ex.ToString();

                synthesisStream = null;
                btnSpeakSSML.IsEnabled = true;
            }

            // 播放音频流
            mediaElement.AutoPlay = true;
            mediaElement.SetSource(synthesisStream, synthesisStream.ContentType);
            mediaElement.Play();
        }

        // 保存 TTS 音频（SSML）
        private async void btnSaveAudio_Click(object sender, RoutedEventArgs e)
        {
            FileSavePicker savePicker = new FileSavePicker();
            savePicker.DefaultFileExtension = ".wav";
            savePicker.FileTypeChoices.Add("Audio file", new List<string>() { ".wav" });

            StorageFile file = await savePicker.PickSaveFileAsync();
            if (file != null)
            {
                btnSaveAudio.IsEnabled = false;

                SpeechSynthesisStream synthesisStream;
                try
                {
                    // 构造一个 SSML 协议文档
                    string ssml = "<speak version=\"1.0\" xmlns=\"http://www.w3.org/2001/10/synthesis\" xml:lang=\"zh-CN\">"; // 中文
                    ssml += "<voice gender=\"male\">"; // 男声
                    ssml += "<prosody rate=\"-50%\">"; // 语速放慢 50%
                    ssml += "TTS 是 Text To Speech 的缩写，即“从文本到语音”，是人机对话的一部分，让机器能够说话。";
                    ssml += "</prosody>";
                    ssml += "</voice>";
                    ssml += "</speak>";

                    // 通过 TTS 引擎将 SSML 协议文档转换成音频流
                    synthesisStream = await _synthesizer.SynthesizeSsmlToStreamAsync(ssml);
                }
                catch (Exception ex)
                {
                    lblMsg.Text = ex.ToString();

                    synthesisStream = null;
                    btnSaveAudio.IsEnabled = true;
                }

                // 保存音频数据到文件
                Windows.Storage.Streams.Buffer buffer = new Windows.Storage.Streams.Buffer(4096);
                IRandomAccessStream writeStream = (IRandomAccessStream)await file.OpenAsync(FileAccessMode.ReadWrite);
                IOutputStream outputStream = writeStream.GetOutputStreamAt(0);
                DataWriter dataWriter = new DataWriter(outputStream);

                while (synthesisStream.Position < synthesisStream.Size)
                {
                    await synthesisStream.ReadAsync(buffer, 4096, InputStreamOptions.None);
                    dataWriter.WriteBuffer(buffer);
                }

                dataWriter.StoreAsync().AsTask().Wait();
                outputStream.FlushAsync().AsTask().Wait();

                btnSaveAudio.IsEnabled = true;
            }
        }
    }
}