sp_enc.h

实现了录音,放音功能!在evc4.0下编译功过,wince5.0下能正常录音,放音,暂停录放音!

/*************************************************************************/
/*                                                                       */
/* Copyright (c) 2000-2004 Linuos Design                                 */
/*                                     领驰设计中心  版权所有 2000-2004  */
/*                                                                       */
/* PROPRIETARY RIGHTS of Linuos Design  are involved in the subject      */
/* matter of this material.  All manufacturing, reproduction, use, and   */
/* sales rights pertaining to this subject matter are governed by the    */
/* license agreement.  The recipient of this software implicitly accepts */ 
/* the terms of the license.                                             */
/* 本软件文档资料是领驰设计中心的资产,任何人士阅读和使用本资料必须获得   */
/* 相应的书面授权,承担保密责任和接受相应的法律约束.                      */
/*                                                                       */
/*************************************************************************/

/*
 * ===================================================================
 *  TS 26.104
 *  REL-5 V5.4.0 2004-03
 *  REL-6 V6.1.0 2004-03
 *  3GPP AMR Floating-point Speech Codec
 * ===================================================================
 *
 */

/*
 * sp_enc.h
 *
 *
 * Project:
 *    AMR Floating-Point Codec
 *
 * Contains:
 *    Defines interface to AMR encoder
 *
 */
#ifndef _SP_ENC_H
#define _SP_ENC_H

/*
 * include files
 */
//#include "typedef.h"
#include "rom_enc.h"


#define ABSTOR( x ) ( ( INT32 )( ( (x)>0 ) ? (x) : -(x) ) )
#define LABSTOR( x ) ( ( INT32 )( ( (x)>0 ) ? (x) : -(x) ) )

/*
* Definition of structures used in encoding process
*/
typedef struct
{
	float y2;
	float y1;
	float x0;
	float x1;
	
}Pre_ProcessState;

typedef struct
{
	float bckr_est[NB_SUBBDS_VAD];   /* background noise estimate */
	float ave_level[NB_SUBBDS_VAD];
	
	
	/* averaged input components for stationary estimation */
	float old_level[NB_SUBBDS_VAD];   /* input levels of the previous frame */
	float sub_level[NB_SUBBDS_VAD];
	
	
	/* input levels calculated at the end of a frame (lookahead) */
	float a_data5[3][2];   /* memory for the filter bank */
	float a_data3[5];   /* memory for the filter bank */
	float best_corr_hp;   /* FIP filtered value */
	
	
	/* counts length of a speech burst incl HO addition */
	float corr_hp_fast;   /* filtered value */
	INT32 vadreg;   /* flags for intermediate VAD decisions */
	INT32 pitch;   /* flags for pitch detection */
	INT32 oldlag_count;
	INT32 oldlag;   /* variables for pitch detection */
	INT32 complex_high;   /* flags for complex detection */
	INT32 complex_low;   /* flags for complex detection */
	INT32 complex_warning;   /* complex background warning */
	INT32 tone;   /* flags for tone detection */
	INT16 burst_count;   /* counts length of a speech burst */
	INT16 hang_count;   /* hangover counter */
	INT16 stat_count;   /* stationary counter */
	INT16 complex_hang_count;   /* complex hangover counter, used by VAD */
	INT16 complex_hang_timer;   /* hangover initiator, used by CAD */
	INT16 speech_vad_decision;   /* final decision */
	INT16 sp_burst_count;
	
	
}vadState;
//#endif
#define DTX_HIST_SIZE 8
#define DTX_ELAPSED_FRAMES_THRESH (24 + 7 -1)
#define DTX_HANG_CONST 7   /* yields eight frames of SP HANGOVER */
typedef struct
{
	float lsp_hist[LP_ORDER * DTX_HIST_SIZE];
	float log_en_hist[DTX_HIST_SIZE];
	INT32 init_lsf_vq_index;
	INT16 hist_ptr;
	INT16 log_en_index;
	INT16 lsp_index[3];
	
	
	/* DTX handler stuff */
	INT16 dtxHangoverCount;
	INT16 decAnaElapsedCount;
	
	
}dtx_encState;
typedef struct
{
	/* gain history */
	float gp[AVG_PITGAN];
	
	
	/* counters */
	INT16 count;
	
	
}tonStabState;
typedef struct
{
	INT32 past_qua_en[4];
	
	
	/* normal MA predictor memory, (contains 20*log10(qua_err)) */
}gc_predState;

typedef struct
{
	float prev_alpha;   /* previous adaptor output, */
	float prev_gc;   /* previous code gain, */
	float ltpg_mem[LTPG_MEM_SIZE];   /* LTP coding gain history, */
	INT16 onset;   /* onset state, */
	
	
	/* (ltpg_mem[0] not used for history) */
}gain_adaptState;
typedef struct
{
	
	float sf0_target_en;
	float sf0_coeff[5];
	INT32 sf0_gcode0_exp;
	INT32 sf0_gcode0_fra;
	INT16 *gain_idx_ptr;
	
	
	gc_predState gc_predSt;
	gc_predState gc_predUncSt;
	gain_adaptState adaptSt;
}gainQuantState;
typedef struct
{
	INT32 T0_prev_subframe;   /* integer pitch lag of previous sub-frame */
	
	
}Pitch_frState;
typedef struct
{
	Pitch_frState pitchSt;
}clLtpState;
typedef struct
{
	float ada_w;
	INT32 old_T0_med;
	INT16 wght_flg;
	
	
}pitchOLWghtState;
typedef struct
{
	float past_rq[LP_ORDER];   /* Past quantized prediction error */
	
	
}Q_plsfState;
typedef struct
{
	/* Past LSPs */
	float lsp_old[LP_ORDER];
	float lsp_old_q[LP_ORDER];
	
	
	/* Quantization state */
	Q_plsfState qSt;
}lspState;
typedef struct
{
	float old_A[LP_ORDER + 1];   /* Last A(z) for case of unstable filter */
	
	
}LevinsonState;
typedef struct
{
	LevinsonState LevinsonSt;
}lpcState;
typedef struct
{
	/* Speech vector */
	float old_speech[SPCH_BUF_SIZE];
	float *speech;
	float *p_window;
	float *p_window_12k2;
	float *new_speech;   /* Global variable */
	
	
	/* Weight speech vector */
	float old_wsp[FRAME_SIZE + PIT_MAX];
	float *wsp;
	
	
	/* OL LTP states */
	INT32 old_lags[5];
	float ol_gain_flg[2];
	
	
	/* Excitation vector */
	float old_exc[FRAME_SIZE + PIT_MAX + INTERPOL_LEN];
	float *exc;
	
	
	/* Zero vector */
	float ai_zero[SUBFRM_SIZE + LP_ORDER_PLUS];
	float *zero;
	
	
	/* Impulse response vector */
	float *h1;
	float hvec[SUBFRM_SIZE * 2];
	
	
	/* Substates */
	lpcState lpcSt;
	lspState lspSt;
	clLtpState clLtpSt;
	gainQuantState gainQuantSt;
	pitchOLWghtState pitchOLWghtSt;
	tonStabState tonStabSt;
	vadState vadSt;
	
	INT32 dtx;
	
	
	dtx_encState dtxEncSt;
	
	/* Filter's memory */
	float mem_syn[LP_ORDER];
	float mem_w0[LP_ORDER];
	float mem_w[LP_ORDER];
	float mem_err[LP_ORDER + SUBFRM_SIZE];
	float *error;
	float sharp;
	
	
}cod_amrState;
typedef struct
{
	cod_amrState cod_amr_state;
	Pre_ProcessState pre_state;
	
	INT32 dtx;
	
	
}Speech_Encode_FrameState;


/*
 * definition of modes for encoder
 */
enum Mode { MR475 = 0,
            MR515,
            MR59,
            MR67,
            MR74,
            MR795,
            MR102,
            MR122,
            MRDTX
};

/*
 * Function prototypes
 */

/*
 * initialize one instance of the speech encoder
 * Stores pointer to filter status struct in *st. This pointer has to
 * be passed to Speech_Encode_Frame in each call.
 * returns 0 on success
 */
void *Speech_Encode_Frame_init (INT32 dtx);
/*
 * reset speech encoder (i.e. set state memory to zero)
 * returns 0 on success
 */
INT32 Speech_Encode_Frame_reset(void *st, INT32 dtx);

/*
 * de-initialize speech encoder (i.e. mcFree status struct)
 * stores NULL in *st
 */
 
//void Speech_Encode_Frame_exit (void **st); 

/*
 * Encodes one speech frame
 * Returns analysis parameters
 */
void Speech_Encode_Frame (void *st, enum Mode mode, INT16 *newSpeech,
                          INT16 *prm, enum Mode *usedMode);

#endif