matlab工具箱TTSBOX源码中文分析

function wav = tts(txt,voice,pace,fs)
%TTS text to speech.
%   TTS (TXT) synthesizes speech from string TXT, and speaks it. The audio
%   format is mono, 16 bit, 16k Hz by default.
%   
%   WAV = TTS(TXT) does not vocalize but output to the variable WAV.
%
%   TTS(TXT,VOICE) uses the specific voice. Use TTS('','List') to see a
%   list of availble voices. Default is the first voice.
%
%   TTS(...,PACE) set the pace of speech to PACE. PACE ranges from 
%   -10 (slowest) to 10 (fastest). Default 0.
%
%   TTS(...,FS) set the sampling rate of the speech to FS kHz. FS must be
%   one of the following: 8000, 11025, 12000, 16000, 22050, 24000, 32000,
%       44100, 48000. Default 16.
%   
%   This function requires the Microsoft Win32 Speech API (SAPI).
%
%   Examples:
%       % Speak the text;
%       tts('I can speak.');
%       % List availble voices;
%       tts('I can speak.','List');
%       % Do not speak out, store the speech in a variable;
%       w = tts('I can speak.',[],-4,44100);
%       wavplay(w,44100);
%
%   See also WAVREAD, WAVWRITE, WAVPLAY.

% Written by Siyi Deng; 12-21-2007;

if ~ispc, error('Microsoft Win32 SAPI is required.'); end
if ~ischar(txt), error('First input must be string.'); end

SV = actxserver('SAPI.SpVoice');%创建一个object对象COM.SAPI_SpVoice
TK = invoke(SV,'GetVoices');%创建一个object对象Interface.928....
if nargin > 1   %判断参数多余一个
 % Select voice;%根据文字选择中文发音对象或者英文发音对象。或这系统中的其他国家语言发音对象
    for k = 0:TK.Count-1
        if strcmpi(voice,TK.Item(k).GetDescription)
            SV.Voice = TK.Item(k);
            break;
        elseif strcmpi(voice,'list')
            disp(TK.Item(k).GetDescription);
        end
    end
    % Set pace;
    if nargin > 2
        if isempty(pace), pace = 0; end %如果pace为空那么默认不加速
        if abs(pace) > 10, pace = sign(pace)*10; end %如果加/减速大于10那么只取左右端的极限       
        SV.Rate = pace;%接口语音速率设置
    end
end

if nargin < 4 || ~ismember(fs,[8000,11025,12000,16000,22050,24000,32000,...
        44100,48000]), fs = 16000; end %如果没有采样频率或者采样频率取值不再规定成员里面，那么使用默认值 16kHz

if nargout > 0  %如果函数有输入参数
   % Output variable;
   MS = actxserver('SAPI.SpMemoryStream');%创建COM.API_SpMemoryStream对象
   MS.Format.Type = sprintf('SAFT%dkHz16BitMono',fix(fs/1000));%设置语音生成频率kHz  fix(fs/1000)取商部分
   SV.AudioOutputStream = MS;  %把生成接口MS生成的语音流<-连接->SVobject输出对象
end

invoke(SV,'Speak',txt);%调用COM的服务对象[SV]使用方法['Speak']处理文本数据[txt]

if nargout > 0 %r如果函数有输入数据
    % Convert uint8 to double precision;
    wav = reshape(double(invoke(MS,'GetData')),2,[])';
%invoke(MS,'GetData')调用返回矩阵信息
%根据数据把数据填成提取[2,:]矩阵先列后行，奇数提取，偶数提取
    wav = (wav(:,2)*256+wav(:,1))/32768;
%2字节提取，归一化计算获得单声道wav文件 2^16=32768 (量化16位)
    wav(wav >= 1) = wav(wav >= 1)-2;
%处理数据后大于等于1的相当于对值为复数的进行处理
%16bit =2字节。小端模式（低字节 高字节）
%存放数据在电脑中用补码表示，
%样例： 05 34 =0 0110100 00000101 显然符号位0为正数（正数三码一样）值为0x(3405)
% 05 80=1 0000000 00000101显然符号位1为负数，值为5-32768
    delete(MS);%释放空间
    clear MS;
end

delete(SV); %释放空间
clear SV TK;
pause(0.2);

end % TTS;