encoder.cpp

mpeg4实现的软件代码

/*
This software module was originally developed by 
        Hang-Seop Lee ([e-mail]), ETRI
        Jung-Chul Lee ([e-mail]), ETRI
and edited by Hang-Seop Lee, Jung-Chul Lee of ETRI, 
in the course of development of the MPEG-4.
This software module is an implementation of a part of one or
more MPEG-4 tools as specified by the MPEG-4.
ISO/IEC gives users of the MPEG-4 free license to this                
software module or modifications thereof for use in hardware
or software products claiming conformance to the MPEG-4.
Those intending to use this software module in hardware or software
products are advised that its use may infringe existing patents.
The original developer of this software module and his/her company,
the subsequent editors and their companies, and ISO/IEC have no
liability for use of this software module or modifications thereof
in an implementation.
Copyright is not released for non MPEG-4 conforming
products. ETRI retains full right to use the code for his/her own
purpose, assign or donate the code to a third party and to
inhibit third parties from using the code for non
MPEG-4 conforming products.
This copyright notice must be included in all copies or
derivative works. Copyright (c) 1997.
*/    

#include "stdafx.h"
#include <stdio.h>
#include <string.h>
#include <stdlib.h>
#include <malloc.h>
#include <sys/stat.h>
#include <ctype.h>
#include <fcntl.h>
#include <windows.h>

	/* For Bit stream */
char YY[2];

int G_Lang_v, G_Gender, G_Gender_v, G_Age, G_Age_v, G_Spch, G_Spch_v;
int G_Proso, G_Proso_d, G_Proso_F0, G_Proso_e, G_Video, G_Lip, G_Trick;

class ttsProsody {
 private:
 public:
    ttsProsody();
    ~ttsProsody();
    char durEnable;
    char f0ContourEnable;
    char energyContourEnable;
    int numberPhonemes;
    int phonemeSymbolsLength;
    char *phonemeSymbols;
    int *durEachPhoneme;
    int *f0ContourEachPhoneme;
    int *energyContourEachPhoneme;
};

ttsProsody::ttsProsody()
{
    numberPhonemes=0;
    phonemeSymbolsLength=0;
    phonemeSymbols=NULL;

    durEnable=0;
    durEachPhoneme=NULL;

    f0ContourEnable=0;
    f0ContourEachPhoneme=NULL;

    energyContourEnable=0;
    energyContourEachPhoneme=NULL;
}

class ttsSentence {
 private:
 public:
    ttsSentence(int i);
    ~ttsSentence();
    int tts_sentence_start_code;
    int ttsSentenceID;
    char silence;
    int silenceDuration;
    char gender;
    char age;
    char speechRate;
    int lengthText;
    char *ttsText;
    ttsProsody *prosody;
    float startTime;
    int sentenceDuration;
    int positionInSentence;
    int offset;
    int numberLipShape;
    int *LipShapeInSentence;
    int *lipShape;
};

ttsSentence::ttsSentence(int i)
{
    tts_sentence_start_code=0x012121212;
    ttsSentenceID=i;
    silence=0;
    silenceDuration=0;
    gender=0;
    age=0;
    speechRate=3;  /* r=[0.7 - 1.6], R=r*10 -3 : default r=1 */
    lengthText=0;
    ttsText=NULL;
    prosody=NULL;
    startTime=(float)0.;
    sentenceDuration=0;
    positionInSentence=0;
    offset=0;
    numberLipShape=0;
    LipShapeInSentence=NULL;
    lipShape=NULL;
}


class ttsSequence {
 private:
    int tts_sequence_start_code;
    int ttsSequenceID;
    char languageCode;
    char genderEnable;
    char ageEnable;
    char speechRateEnable;
    char prosodyEnable;
    char videoEnable;
    char lipShapeEnable;
    char trickModeEnable;
    ttsSentence *sentence;
    int itime;
    int iframe;
 public:
    ttsSequence(int i);
    ~ttsSequence();
    void init();
    void set_LangEn(int i) { languageCode=i; }
    void set_genderEn(int i) { genderEnable=i; }
    void set_ageEn(int i) { ageEnable=i; }
    void set_spRateEn(int i) { speechRateEnable=i; }
    void set_prosodyEn(int i) { prosodyEnable=i; }
    void set_videoEn(int i) { videoEnable=i; }
    void set_lipEn(int i) { lipShapeEnable=i; }
    void set_trickEn(int i) { trickModeEnable=i; }

    int get_LangEn() { return (int)languageCode; }
    int get_genderEn() { return (int)genderEnable; }
    int get_ageEn() { return (int)ageEnable; }
    int get_spRateEn() { return (int)speechRateEnable; }
    int get_prosodyEn() { return (int)prosodyEnable; }
    int get_videoEn() { return (int)videoEnable; }
    int get_lipEn() { return (int)lipShapeEnable; }
    int get_trickEn() { return (int)trickModeEnable; }

    void Sntinit();
    void AddSentence(ttsSentence *ttsSnt) { sentence=ttsSnt; }

    void set_gender(int i) { sentence->gender=i; }
    void set_age(int i) { sentence->age=i; }
    void set_spRate(int i) { sentence->speechRate=i; }
    void set_durEn(int i) { sentence->prosody->durEnable=i; };
    void set_f0En(int i) { sentence->prosody->f0ContourEnable=i; };
    void set_enEn(int i) { sentence->prosody->energyContourEnable=i; };

    int get_gender() { return (int)sentence->gender; }
    int get_age() { return (int)sentence->age; }
    int get_spRate() { return (int)sentence->speechRate; }
    int get_durEn() { return (int)sentence->prosody->durEnable; };
    int get_f0En() { return (int)sentence->prosody->f0ContourEnable; };
    int get_enEn() { return (int)sentence->prosody->energyContourEnable; };

    void put_text(char *Text);
    int get_text(char *Text) 
	{ strcpy(Text,sentence->ttsText); return (int)sentence->lengthText; }
    void put_prosody(int Nphone, short Sphone[][6],short Dur[],short F0[][3],short En[][3]);
    void put_position(int itime, float stime, int dur);
    void put_lip(FILE *fp);
    void save_ttsSequence(FILE *fout);
    void save_ttsSentence(FILE *fout);
    void clear_ttsSentence();
    void data_write(char *A, int nbits, int n, char *B, FILE *fout);
    void data_write(short *A, int nbits, int n, char *B, FILE *fout);
    void data_write(int *A, int nbits, int n, char *B, FILE *fout);
};

ttsSequence::ttsSequence(int i)
{
    tts_sequence_start_code=0x000000011;
    ttsSequenceID=i;
    languageCode=0;
    genderEnable=0;
    ageEnable=0;
    speechRateEnable=0;
    prosodyEnable=0;
    videoEnable=0;
    lipShapeEnable=0;
    trickModeEnable=0;
    sentence=NULL;
    itime=0;
    iframe=0;
}


void ttsSequence::init()
{
    set_LangEn(G_Lang_v);

    set_genderEn(G_Gender);

    set_ageEn(G_Age);

    set_spRateEn(G_Spch);

    set_prosodyEn(G_Proso);

    set_videoEn(G_Video);

    set_lipEn(G_Lip);

    set_trickEn(G_Trick);
}


void ttsSequence::Sntinit()
{
    if(genderEnable) {
    	set_gender(G_Gender_v);
	}

    if(ageEnable) {
    	set_age(G_Age_v);
	}

    if(speechRateEnable) {
    	set_spRate(G_Spch_v);
	}

    if(prosodyEnable) {
    	ttsProsody *prosody = new ttsProsody();
    	sentence->prosody=prosody;
    	set_durEn(G_Proso_d);
    	set_f0En(G_Proso_F0);
    	set_enEn(G_Proso_e);
	}
}

void ttsSequence::put_text(char *Text) 
{ 
    sentence->lengthText=strlen(Text); 
    sentence->ttsText=(char *)malloc(strlen(Text)+2);
    strcpy(sentence->ttsText,Text); 
}

void ttsSequence::put_prosody(int Nphone, short Sphone[][6],short Dur[],short F0[][3],short En[][3])
{ 
    int i, j, k; 
    ttsProsody *p=sentence->prosody;

    p->numberPhonemes=Nphone;
    for(i=k=0; i<Nphone; i++) 
	for(j=0; j<6; j++, k++) if(Sphone[i][j]==0) break;

    i=p->phonemeSymbolsLength=k*2;
    p->phonemeSymbols=(char *)malloc(i+2);
    for(i=k=0; i<Nphone; i++) {
	for(j=0; j<6; j++) {
	    if(Sphone[i][j]==0) break;
	    p->phonemeSymbols[k++]=(Sphone[i][j] >> 8) & 0xFF;
	    p->phonemeSymbols[k++]=(Sphone[i][j] & 0xFF);
	}   }

    if(get_durEn()) {
    	p->durEachPhoneme=(int *)malloc(Nphone*sizeof(int));
    	for(i=0; i<Nphone; i++) p->durEachPhoneme[i]=Dur[i];
	}

    if(get_f0En()) {
    	p->f0ContourEachPhoneme=(int *)malloc(Nphone*3*sizeof(int));
    	for(i=0; i<Nphone; i++) 
    	    for(j=0; j<3; j++) p->f0ContourEachPhoneme[i*3+j]=F0[i][j];
	}

    if(get_enEn()) {
        p->f0ContourEachPhoneme=(int *)malloc(Nphone*3*sizeof(int));
        for(i=0; i<Nphone; i++) 
    	    for(j=0; j<3; j++) p->f0ContourEachPhoneme[i*3+j]=F0[i][j];
	}

}

void ttsSequence::put_position(int t0, float stime, int dur)
{
    int i;

    sentence->startTime=stime;
    sentence->sentenceDuration=dur;
    sentence->positionInSentence=0;
    itime=t0;
    i=(int)(stime*1000); sentence->offset=i%itime;
}

void ttsSequence::save_ttsSequence(FILE *fout)
{
    int i; char c;

    data_write(&tts_sequence_start_code,32,1,YY,fout);
    data_write(&ttsSequenceID,5,1,YY,fout);
    data_write(&languageCode,10,1,YY,fout);
    data_write(&genderEnable,1,1,YY,fout);
    data_write(&ageEnable,1,1,YY,fout);
    data_write(&speechRateEnable,1,1,YY,fout);