/* ** Copyright 2003-2010, VisualOn, Inc. ** ** Licensed under the Apache License, Version 2.0 (the "License"); ** you may not use this file except in compliance with the License. ** You may obtain a copy of the License at ** ** http://www.apache.org/licenses/LICENSE-2.0 ** ** Unless required by applicable law or agreed to in writing, software ** distributed under the License is distributed on an "AS IS" BASIS, ** WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. ** See the License for the specific language governing permissions and ** limitations under the License. */ /*********************************************************************** * File: voAMRWBEnc.c * * * * Description: Performs the main encoder routine * * Fixed-point C simulation of AMR WB ACELP coding * * algorithm with 20 msspeech frames for * * wideband speech signals. * * * ************************************************************************/ #include #include #include "typedef.h" #include "basic_op.h" #include "oper_32b.h" #include "math_op.h" #include "cnst.h" #include "acelp.h" #include "cod_main.h" #include "bits.h" #include "main.h" #include "voAMRWB.h" #include "mem_align.h" #include "cmnMemory.h" #ifdef __cplusplus extern "C" { #endif /* LPC interpolation coef {0.45, 0.8, 0.96, 1.0}; in Q15 */ static Word16 interpol_frac[NB_SUBFR] = {14746, 26214, 31457, 32767}; /* isp tables for initialization */ static Word16 isp_init[M] = { 32138, 30274, 27246, 23170, 18205, 12540, 6393, 0, -6393, -12540, -18205, -23170, -27246, -30274, -32138, 1475 }; static Word16 isf_init[M] = { 1024, 2048, 3072, 4096, 5120, 6144, 7168, 8192, 9216, 10240, 11264, 12288, 13312, 14336, 15360, 3840 }; /* High Band encoding */ static const Word16 HP_gain[16] = { 3624, 4673, 5597, 6479, 7425, 8378, 9324, 10264, 11210, 12206, 13391, 14844, 16770, 19655, 24289, 32728 }; /* Private function declaration */ static Word16 synthesis( Word16 Aq[], /* A(z) : quantized Az */ Word16 exc[], /* (i) : excitation at 12kHz */ Word16 Q_new, /* (i) : scaling performed on exc */ Word16 synth16k[], /* (o) : 16kHz synthesis signal */ Coder_State * st /* (i/o) : State structure */ ); /* Codec some parameters initialization */ void Reset_encoder(void *st, Word16 reset_all) { Word16 i; Coder_State *cod_state; cod_state = (Coder_State *) st; Set_zero(cod_state->old_exc, PIT_MAX + L_INTERPOL); Set_zero(cod_state->mem_syn, M); Set_zero(cod_state->past_isfq, M); cod_state->mem_w0 = 0; cod_state->tilt_code = 0; cod_state->first_frame = 1; Init_gp_clip(cod_state->gp_clip); cod_state->L_gc_thres = 0; if (reset_all != 0) { /* Static vectors to zero */ Set_zero(cod_state->old_speech, L_TOTAL - L_FRAME); Set_zero(cod_state->old_wsp, (PIT_MAX / OPL_DECIM)); Set_zero(cod_state->mem_decim2, 3); /* routines initialization */ Init_Decim_12k8(cod_state->mem_decim); Init_HP50_12k8(cod_state->mem_sig_in); Init_Levinson(cod_state->mem_levinson); Init_Q_gain2(cod_state->qua_gain); Init_Hp_wsp(cod_state->hp_wsp_mem); /* isp initialization */ Copy(isp_init, cod_state->ispold, M); Copy(isp_init, cod_state->ispold_q, M); /* variable initialization */ cod_state->mem_preemph = 0; cod_state->mem_wsp = 0; cod_state->Q_old = 15; cod_state->Q_max[0] = 15; cod_state->Q_max[1] = 15; cod_state->old_wsp_max = 0; cod_state->old_wsp_shift = 0; /* pitch ol initialization */ cod_state->old_T0_med = 40; cod_state->ol_gain = 0; cod_state->ada_w = 0; cod_state->ol_wght_flg = 0; for (i = 0; i < 5; i++) { cod_state->old_ol_lag[i] = 40; } Set_zero(cod_state->old_hp_wsp, (L_FRAME / 2) / OPL_DECIM + (PIT_MAX / OPL_DECIM)); Set_zero(cod_state->mem_syn_hf, M); Set_zero(cod_state->mem_syn_hi, M); Set_zero(cod_state->mem_syn_lo, M); Init_HP50_12k8(cod_state->mem_sig_out); Init_Filt_6k_7k(cod_state->mem_hf); Init_HP400_12k8(cod_state->mem_hp400); Copy(isf_init, cod_state->isfold, M); cod_state->mem_deemph = 0; cod_state->seed2 = 21845; Init_Filt_6k_7k(cod_state->mem_hf2); cod_state->gain_alpha = 32767; cod_state->vad_hist = 0; wb_vad_reset(cod_state->vadSt); dtx_enc_reset(cod_state->dtx_encSt, isf_init); } return; } /*-----------------------------------------------------------------* * Funtion coder * * ~~~~~ * * ->Main coder routine. * * * *-----------------------------------------------------------------*/ void coder( Word16 * mode, /* input : used mode */ Word16 speech16k[], /* input : 320 new speech samples (at 16 kHz) */ Word16 prms[], /* output: output parameters */ Word16 * ser_size, /* output: bit rate of the used mode */ void *spe_state, /* i/o : State structure */ Word16 allow_dtx /* input : DTX ON/OFF */ ) { /* Coder states */ Coder_State *st; /* Speech vector */ Word16 old_speech[L_TOTAL]; Word16 *new_speech, *speech, *p_window; /* Weighted speech vector */ Word16 old_wsp[L_FRAME + (PIT_MAX / OPL_DECIM)]; Word16 *wsp; /* Excitation vector */ Word16 old_exc[(L_FRAME + 1) + PIT_MAX + L_INTERPOL]; Word16 *exc; /* LPC coefficients */ Word16 r_h[M + 1], r_l[M + 1]; /* Autocorrelations of windowed speech */ Word16 rc[M]; /* Reflection coefficients. */ Word16 Ap[M + 1]; /* A(z) with spectral expansion */ Word16 ispnew[M]; /* immittance spectral pairs at 4nd sfr */ Word16 ispnew_q[M]; /* quantized ISPs at 4nd subframe */ Word16 isf[M]; /* ISF (frequency domain) at 4nd sfr */ Word16 *p_A, *p_Aq; /* ptr to A(z) for the 4 subframes */ Word16 A[NB_SUBFR * (M + 1)]; /* A(z) unquantized for the 4 subframes */ Word16 Aq[NB_SUBFR * (M + 1)]; /* A(z) quantized for the 4 subframes */ /* Other vectors */ Word16 xn[L_SUBFR]; /* Target vector for pitch search */ Word16 xn2[L_SUBFR]; /* Target vector for codebook search */ Word16 dn[L_SUBFR]; /* Correlation between xn2 and h1 */ Word16 cn[L_SUBFR]; /* Target vector in residual domain */ Word16 h1[L_SUBFR]; /* Impulse response vector */ Word16 h2[L_SUBFR]; /* Impulse response vector */ Word16 code[L_SUBFR]; /* Fixed codebook excitation */ Word16 y1[L_SUBFR]; /* Filtered adaptive excitation */ Word16 y2[L_SUBFR]; /* Filtered adaptive excitation */ Word16 error[M + L_SUBFR]; /* error of quantization */ Word16 synth[L_SUBFR]; /* 12.8kHz synthesis vector */ Word16 exc2[L_FRAME]; /* excitation vector */ Word16 buf[L_FRAME]; /* VAD buffer */ /* Scalars */ Word32 i, j, i_subfr, select, pit_flag, clip_gain, vad_flag; Word16 codec_mode; Word16 T_op, T_op2, T0, T0_min, T0_max, T0_frac, index; Word16 gain_pit, gain_code, g_coeff[4], g_coeff2[4]; Word16 tmp, gain1, gain2, exp, Q_new, mu, shift, max; Word16 voice_fac; Word16 indice[8]; Word32 L_tmp, L_gain_code, L_max, L_tmp1; Word16 code2[L_SUBFR]; /* Fixed codebook excitation */ Word16 stab_fac, fac, gain_code_lo; Word16 corr_gain; Word16 *vo_p0, *vo_p1, *vo_p2, *vo_p3; st = (Coder_State *) spe_state; *ser_size = nb_of_bits[*mode]; codec_mode = *mode; /*--------------------------------------------------------------------------* * Initialize pointers to speech vector. * * * * * * |-------|-------|-------|-------|-------|-------| * * past sp sf1 sf2 sf3 sf4 L_NEXT * * <------- Total speech buffer (L_TOTAL) ------> * * old_speech * * <------- LPC analysis window (L_WINDOW) ------> * * | <-- present frame (L_FRAME) ----> * * p_window | <----- new speech (L_FRAME) ----> * * | | * * speech | * * new_speech * *--------------------------------------------------------------------------*/ new_speech = old_speech + L_TOTAL - L_FRAME - L_FILT; /* New speech */ speech = old_speech + L_TOTAL - L_FRAME - L_NEXT; /* Present frame */ p_window = old_speech + L_TOTAL - L_WINDOW; exc = old_exc + PIT_MAX + L_INTERPOL; wsp = old_wsp + (PIT_MAX / OPL_DECIM); /* copy coder memory state into working space */ Copy(st->old_speech, old_speech, L_TOTAL - L_FRAME); Copy(st->old_wsp, old_wsp, PIT_MAX / OPL_DECIM); Copy(st->old_exc, old_exc, PIT_MAX + L_INTERPOL); /*---------------------------------------------------------------* * Down sampling signal from 16kHz to 12.8kHz * * -> The signal is extended by L_FILT samples (padded to zero) * * to avoid additional delay (L_FILT samples) in the coder. * * The last L_FILT samples are approximated after decimation and * * are used (and windowed) only in autocorrelations. * *---------------------------------------------------------------*/ Decim_12k8(speech16k, L_FRAME16k, new_speech, st->mem_decim); /* last L_FILT samples for autocorrelation window */ Copy(st->mem_decim, code, 2 * L_FILT16k); Set_zero(error, L_FILT16k); /* set next sample to zero */ Decim_12k8(error, L_FILT16k, new_speech + L_FRAME, code); /*---------------------------------------------------------------* * Perform 50Hz HP filtering of input signal. * *---------------------------------------------------------------*/ HP50_12k8(new_speech, L_FRAME, st->mem_sig_in); /* last L_FILT samples for autocorrelation window */ Copy(st->mem_sig_in, code, 6); HP50_12k8(new_speech + L_FRAME, L_FILT, code); /*---------------------------------------------------------------* * Perform fixed preemphasis through 1 - g z^-1 * * Scale signal to get maximum of precision in filtering * *---------------------------------------------------------------*/ mu = PREEMPH_FAC >> 1; /* Q15 --> Q14 */ /* get max of new preemphased samples (L_FRAME+L_FILT) */ L_tmp = new_speech[0] << 15; L_tmp -= (st->mem_preemph * mu)<<1; L_max = L_abs(L_tmp); for (i = 1; i < L_FRAME + L_FILT; i++) { L_tmp = new_speech[i] << 15; L_tmp -= (new_speech[i - 1] * mu)<<1; L_tmp = L_abs(L_tmp); if(L_tmp > L_max) { L_max = L_tmp; } } /* get scaling factor for new and previous samples */ /* limit scaling to Q_MAX to keep dynamic for ringing in low signal */ /* limit scaling to Q_MAX also to avoid a[0]<1 in syn_filt_32 */ tmp = extract_h(L_max); if (tmp == 0) { shift = Q_MAX; } else { shift = norm_s(tmp) - 1; if (shift < 0) { shift = 0; } if (shift > Q_MAX) { shift = Q_MAX; } } Q_new = shift; if (Q_new > st->Q_max[0]) { Q_new = st->Q_max[0]; } if (Q_new > st->Q_max[1]) { Q_new = st->Q_max[1]; } exp = (Q_new - st->Q_old); st->Q_old = Q_new; st->Q_max[1] = st->Q_max[0]; st->Q_max[0] = shift; /* preemphasis with scaling (L_FRAME+L_FILT) */ tmp = new_speech[L_FRAME - 1]; for (i = L_FRAME + L_FILT - 1; i > 0; i--) { L_tmp = new_speech[i] << 15; L_tmp -= (new_speech[i - 1] * mu)<<1; L_tmp = (L_tmp << Q_new); new_speech[i] = vo_round(L_tmp); } L_tmp = new_speech[0] << 15; L_tmp -= (st->mem_preemph * mu)<<1; L_tmp = (L_tmp << Q_new); new_speech[0] = vo_round(L_tmp); st->mem_preemph = tmp; /* scale previous samples and memory */ Scale_sig(old_speech, L_TOTAL - L_FRAME - L_FILT, exp); Scale_sig(old_exc, PIT_MAX + L_INTERPOL, exp); Scale_sig(st->mem_syn, M, exp); Scale_sig(st->mem_decim2, 3, exp); Scale_sig(&(st->mem_wsp), 1, exp); Scale_sig(&(st->mem_w0), 1, exp); /*------------------------------------------------------------------------* * Call VAD * * Preemphesis scale down signal in low frequency and keep dynamic in HF.* * Vad work slightly in futur (new_speech = speech + L_NEXT - L_FILT). * *------------------------------------------------------------------------*/ Copy(new_speech, buf, L_FRAME); #ifdef ASM_OPT /* asm optimization branch */ Scale_sig_opt(buf, L_FRAME, 1 - Q_new); #else Scale_sig(buf, L_FRAME, 1 - Q_new); #endif vad_flag = wb_vad(st->vadSt, buf); /* Voice Activity Detection */ if (vad_flag == 0) { st->vad_hist = (st->vad_hist + 1); } else { st->vad_hist = 0; } /* DTX processing */ if (allow_dtx != 0) { /* Note that mode may change here */ tx_dtx_handler(st->dtx_encSt, vad_flag, mode); *ser_size = nb_of_bits[*mode]; } if(*mode != MRDTX) { Parm_serial(vad_flag, 1, &prms); } /*------------------------------------------------------------------------* * Perform LPC analysis * * ~~~~~~~~~~~~~~~~~~~~ * * - autocorrelation + lag windowing * * - Levinson-durbin algorithm to find a[] * * - convert a[] to isp[] * * - convert isp[] to isf[] for quantization * * - quantize and code the isf[] * * - convert isf[] to isp[] for interpolation * * - find the interpolated ISPs and convert to a[] for the 4 subframes * *------------------------------------------------------------------------*/ /* LP analysis centered at 4nd subframe */ Autocorr(p_window, M, r_h, r_l); /* Autocorrelations */ Lag_window(r_h, r_l); /* Lag windowing */ Levinson(r_h, r_l, A, rc, st->mem_levinson); /* Levinson Durbin */ Az_isp(A, ispnew, st->ispold); /* From A(z) to ISP */ /* Find the interpolated ISPs and convert to a[] for all subframes */ Int_isp(st->ispold, ispnew, interpol_frac, A); /* update ispold[] for the next frame */ Copy(ispnew, st->ispold, M); /* Convert ISPs to frequency domain 0..6400 */ Isp_isf(ispnew, isf, M); /* check resonance for pitch clipping algorithm */ Gp_clip_test_isf(isf, st->gp_clip); /*----------------------------------------------------------------------* * Perform PITCH_OL analysis * * ~~~~~~~~~~~~~~~~~~~~~~~~~ * * - Find the residual res[] for the whole speech frame * * - Find the weighted input speech wsp[] for the whole speech frame * * - scale wsp[] to avoid overflow in pitch estimation * * - Find open loop pitch lag for whole speech frame * *----------------------------------------------------------------------*/ p_A = A; for (i_subfr = 0; i_subfr < L_FRAME; i_subfr += L_SUBFR) { /* Weighting of LPC coefficients */ Weight_a(p_A, Ap, GAMMA1, M); #ifdef ASM_OPT /* asm optimization branch */ Residu_opt(Ap, &speech[i_subfr], &wsp[i_subfr], L_SUBFR); #else Residu(Ap, &speech[i_subfr], &wsp[i_subfr], L_SUBFR); #endif p_A += (M + 1); } Deemph2(wsp, TILT_FAC, L_FRAME, &(st->mem_wsp)); /* find maximum value on wsp[] for 12 bits scaling */ max = 0; for (i = 0; i < L_FRAME; i++) { tmp = abs_s(wsp[i]); if(tmp > max) { max = tmp; } } tmp = st->old_wsp_max; if(max > tmp) { tmp = max; /* tmp = max(wsp_max, old_wsp_max) */ } st->old_wsp_max = max; shift = norm_s(tmp) - 3; if (shift > 0) { shift = 0; /* shift = 0..-3 */ } /* decimation of wsp[] to search pitch in LF and to reduce complexity */ LP_Decim2(wsp, L_FRAME, st->mem_decim2); /* scale wsp[] in 12 bits to avoid overflow */ #ifdef ASM_OPT /* asm optimization branch */ Scale_sig_opt(wsp, L_FRAME / OPL_DECIM, shift); #else Scale_sig(wsp, L_FRAME / OPL_DECIM, shift); #endif /* scale old_wsp (warning: exp must be Q_new-Q_old) */ exp = exp + (shift - st->old_wsp_shift); st->old_wsp_shift = shift; Scale_sig(old_wsp, PIT_MAX / OPL_DECIM, exp); Scale_sig(st->old_hp_wsp, PIT_MAX / OPL_DECIM, exp); scale_mem_Hp_wsp(st->hp_wsp_mem, exp); /* Find open loop pitch lag for whole speech frame */ if(*ser_size == NBBITS_7k) { /* Find open loop pitch lag for whole speech frame */ T_op = Pitch_med_ol(wsp, st, L_FRAME / OPL_DECIM); } else { /* Find open loop pitch lag for first 1/2 frame */ T_op = Pitch_med_ol(wsp, st, (L_FRAME/2) / OPL_DECIM); } if(st->ol_gain > 19661) /* 0.6 in Q15 */ { st->old_T0_med = Med_olag(T_op, st->old_ol_lag); st->ada_w = 32767; } else { st->ada_w = vo_mult(st->ada_w, 29491); } if(st->ada_w < 26214) st->ol_wght_flg = 0; else st->ol_wght_flg = 1; wb_vad_tone_detection(st->vadSt, st->ol_gain); T_op *= OPL_DECIM; if(*ser_size != NBBITS_7k) { /* Find open loop pitch lag for second 1/2 frame */ T_op2 = Pitch_med_ol(wsp + ((L_FRAME / 2) / OPL_DECIM), st, (L_FRAME/2) / OPL_DECIM); if(st->ol_gain > 19661) /* 0.6 in Q15 */ { st->old_T0_med = Med_olag(T_op2, st->old_ol_lag); st->ada_w = 32767; } else { st->ada_w = mult(st->ada_w, 29491); } if(st->ada_w < 26214) st->ol_wght_flg = 0; else st->ol_wght_flg = 1; wb_vad_tone_detection(st->vadSt, st->ol_gain); T_op2 *= OPL_DECIM; } else { T_op2 = T_op; } /*----------------------------------------------------------------------* * DTX-CNG * *----------------------------------------------------------------------*/ if(*mode == MRDTX) /* CNG mode */ { /* Buffer isf's and energy */ #ifdef ASM_OPT /* asm optimization branch */ Residu_opt(&A[3 * (M + 1)], speech, exc, L_FRAME); #else Residu(&A[3 * (M + 1)], speech, exc, L_FRAME); #endif for (i = 0; i < L_FRAME; i++) { exc2[i] = shr(exc[i], Q_new); } L_tmp = 0; for (i = 0; i < L_FRAME; i++) L_tmp += (exc2[i] * exc2[i])<<1; L_tmp >>= 1; dtx_buffer(st->dtx_encSt, isf, L_tmp, codec_mode); /* Quantize and code the ISFs */ dtx_enc(st->dtx_encSt, isf, exc2, &prms); /* Convert ISFs to the cosine domain */ Isf_isp(isf, ispnew_q, M); Isp_Az(ispnew_q, Aq, M, 0); for (i_subfr = 0; i_subfr < L_FRAME; i_subfr += L_SUBFR) { corr_gain = synthesis(Aq, &exc2[i_subfr], 0, &speech16k[i_subfr * 5 / 4], st); } Copy(isf, st->isfold, M); /* reset speech coder memories */ Reset_encoder(st, 0); /*--------------------------------------------------* * Update signal for next frame. * * -> save past of speech[] and wsp[]. * *--------------------------------------------------*/ Copy(&old_speech[L_FRAME], st->old_speech, L_TOTAL - L_FRAME); Copy(&old_wsp[L_FRAME / OPL_DECIM], st->old_wsp, PIT_MAX / OPL_DECIM); return; } /*----------------------------------------------------------------------* * ACELP * *----------------------------------------------------------------------*/ /* Quantize and code the ISFs */ if (*ser_size <= NBBITS_7k) { Qpisf_2s_36b(isf, isf, st->past_isfq, indice, 4); Parm_serial(indice[0], 8, &prms); Parm_serial(indice[1], 8, &prms); Parm_serial(indice[2], 7, &prms); Parm_serial(indice[3], 7, &prms); Parm_serial(indice[4], 6, &prms); } else { Qpisf_2s_46b(isf, isf, st->past_isfq, indice, 4); Parm_serial(indice[0], 8, &prms); Parm_serial(indice[1], 8, &prms); Parm_serial(indice[2], 6, &prms); Parm_serial(indice[3], 7, &prms); Parm_serial(indice[4], 7, &prms); Parm_serial(indice[5], 5, &prms); Parm_serial(indice[6], 5, &prms); } /* Check stability on isf : distance between old isf and current isf */ L_tmp = 0; for (i = 0; i < M - 1; i++) { tmp = vo_sub(isf[i], st->isfold[i]); L_tmp += (tmp * tmp)<<1; } tmp = extract_h(L_shl2(L_tmp, 8)); tmp = vo_mult(tmp, 26214); /* tmp = L_tmp*0.8/256 */ tmp = vo_sub(20480, tmp); /* 1.25 - tmp (in Q14) */ stab_fac = shl(tmp, 1); if (stab_fac < 0) { stab_fac = 0; } Copy(isf, st->isfold, M); /* Convert ISFs to the cosine domain */ Isf_isp(isf, ispnew_q, M); if (st->first_frame != 0) { st->first_frame = 0; Copy(ispnew_q, st->ispold_q, M); } /* Find the interpolated ISPs and convert to a[] for all subframes */ Int_isp(st->ispold_q, ispnew_q, interpol_frac, Aq); /* update ispold[] for the next frame */ Copy(ispnew_q, st->ispold_q, M); p_Aq = Aq; for (i_subfr = 0; i_subfr < L_FRAME; i_subfr += L_SUBFR) { #ifdef ASM_OPT /* asm optimization branch */ Residu_opt(p_Aq, &speech[i_subfr], &exc[i_subfr], L_SUBFR); #else Residu(p_Aq, &speech[i_subfr], &exc[i_subfr], L_SUBFR); #endif p_Aq += (M + 1); } /* Buffer isf's and energy for dtx on non-speech frame */ if (vad_flag == 0) { for (i = 0; i < L_FRAME; i++) { exc2[i] = exc[i] >> Q_new; } L_tmp = 0; for (i = 0; i < L_FRAME; i++) L_tmp += (exc2[i] * exc2[i])<<1; L_tmp >>= 1; dtx_buffer(st->dtx_encSt, isf, L_tmp, codec_mode); } /* range for closed loop pitch search in 1st subframe */ T0_min = T_op - 8; if (T0_min < PIT_MIN) { T0_min = PIT_MIN; } T0_max = (T0_min + 15); if(T0_max > PIT_MAX) { T0_max = PIT_MAX; T0_min = T0_max - 15; } /*------------------------------------------------------------------------* * Loop for every subframe in the analysis frame * *------------------------------------------------------------------------* * To find the pitch and innovation parameters. The subframe size is * * L_SUBFR and the loop is repeated L_FRAME/L_SUBFR times. * * - compute the target signal for pitch search * * - compute impulse response of weighted synthesis filter (h1[]) * * - find the closed-loop pitch parameters * * - encode the pitch dealy * * - find 2 lt prediction (with / without LP filter for lt pred) * * - find 2 pitch gains and choose the best lt prediction. * * - find target vector for codebook search * * - update the impulse response h1[] for codebook search * * - correlation between target vector and impulse response * * - codebook search and encoding * * - VQ of pitch and codebook gains * * - find voicing factor and tilt of code for next subframe. * * - update states of weighting filter * * - find excitation and synthesis speech * *------------------------------------------------------------------------*/ p_A = A; p_Aq = Aq; for (i_subfr = 0; i_subfr < L_FRAME; i_subfr += L_SUBFR) { pit_flag = i_subfr; if ((i_subfr == 2 * L_SUBFR) && (*ser_size > NBBITS_7k)) { pit_flag = 0; /* range for closed loop pitch search in 3rd subframe */ T0_min = (T_op2 - 8); if (T0_min < PIT_MIN) { T0_min = PIT_MIN; } T0_max = (T0_min + 15); if (T0_max > PIT_MAX) { T0_max = PIT_MAX; T0_min = (T0_max - 15); } } /*-----------------------------------------------------------------------* * * * Find the target vector for pitch search: * * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ * * * * |------| res[n] * * speech[n]---| A(z) |-------- * * |------| | |--------| error[n] |------| * * zero -- (-)--| 1/A(z) |-----------| W(z) |-- target * * exc |--------| |------| * * * * Instead of subtracting the zero-input response of filters from * * the weighted input speech, the above configuration is used to * * compute the target vector. * * * *-----------------------------------------------------------------------*/ for (i = 0; i < M; i++) { error[i] = vo_sub(speech[i + i_subfr - M], st->mem_syn[i]); } #ifdef ASM_OPT /* asm optimization branch */ Residu_opt(p_Aq, &speech[i_subfr], &exc[i_subfr], L_SUBFR); #else Residu(p_Aq, &speech[i_subfr], &exc[i_subfr], L_SUBFR); #endif Syn_filt(p_Aq, &exc[i_subfr], error + M, L_SUBFR, error, 0); Weight_a(p_A, Ap, GAMMA1, M); #ifdef ASM_OPT /* asm optimization branch */ Residu_opt(Ap, error + M, xn, L_SUBFR); #else Residu(Ap, error + M, xn, L_SUBFR); #endif Deemph2(xn, TILT_FAC, L_SUBFR, &(st->mem_w0)); /*----------------------------------------------------------------------* * Find approx. target in residual domain "cn[]" for inovation search. * *----------------------------------------------------------------------*/ /* first half: xn[] --> cn[] */ Set_zero(code, M); Copy(xn, code + M, L_SUBFR / 2); tmp = 0; Preemph2(code + M, TILT_FAC, L_SUBFR / 2, &tmp); Weight_a(p_A, Ap, GAMMA1, M); Syn_filt(Ap,code + M, code + M, L_SUBFR / 2, code, 0); #ifdef ASM_OPT /* asm optimization branch */ Residu_opt(p_Aq,code + M, cn, L_SUBFR / 2); #else Residu(p_Aq,code + M, cn, L_SUBFR / 2); #endif /* second half: res[] --> cn[] (approximated and faster) */ Copy(&exc[i_subfr + (L_SUBFR / 2)], cn + (L_SUBFR / 2), L_SUBFR / 2); /*---------------------------------------------------------------* * Compute impulse response, h1[], of weighted synthesis filter * *---------------------------------------------------------------*/ Set_zero(error, M + L_SUBFR); Weight_a(p_A, error + M, GAMMA1, M); vo_p0 = error+M; vo_p3 = h1; for (i = 0; i < L_SUBFR; i++) { L_tmp = *vo_p0 << 14; /* x4 (Q12 to Q14) */ vo_p1 = p_Aq + 1; vo_p2 = vo_p0-1; for (j = 1; j <= M/4; j++) { L_tmp -= *vo_p1++ * *vo_p2--; L_tmp -= *vo_p1++ * *vo_p2--; L_tmp -= *vo_p1++ * *vo_p2--; L_tmp -= *vo_p1++ * *vo_p2--; } *vo_p3++ = *vo_p0++ = vo_round((L_tmp <<4)); } /* deemph without division by 2 -> Q14 to Q15 */ tmp = 0; Deemph2(h1, TILT_FAC, L_SUBFR, &tmp); /* h1 in Q14 */ /* h2 in Q12 for codebook search */ Copy(h1, h2, L_SUBFR); /*---------------------------------------------------------------* * scale xn[] and h1[] to avoid overflow in dot_product12() * *---------------------------------------------------------------*/ #ifdef ASM_OPT /* asm optimization branch */ Scale_sig_opt(h2, L_SUBFR, -2); Scale_sig_opt(xn, L_SUBFR, shift); /* scaling of xn[] to limit dynamic at 12 bits */ Scale_sig_opt(h1, L_SUBFR, 1 + shift); /* set h1[] in Q15 with scaling for convolution */ #else Scale_sig(h2, L_SUBFR, -2); Scale_sig(xn, L_SUBFR, shift); /* scaling of xn[] to limit dynamic at 12 bits */ Scale_sig(h1, L_SUBFR, 1 + shift); /* set h1[] in Q15 with scaling for convolution */ #endif /*----------------------------------------------------------------------* * Closed-loop fractional pitch search * *----------------------------------------------------------------------*/ /* find closed loop fractional pitch lag */ if(*ser_size <= NBBITS_9k) { T0 = Pitch_fr4(&exc[i_subfr], xn, h1, T0_min, T0_max, &T0_frac, pit_flag, PIT_MIN, PIT_FR1_8b, L_SUBFR); /* encode pitch lag */ if (pit_flag == 0) /* if 1st/3rd subframe */ { /*--------------------------------------------------------------* * The pitch range for the 1st/3rd subframe is encoded with * * 8 bits and is divided as follows: * * PIT_MIN to PIT_FR1-1 resolution 1/2 (frac = 0 or 2) * * PIT_FR1 to PIT_MAX resolution 1 (frac = 0) * *--------------------------------------------------------------*/ if (T0 < PIT_FR1_8b) { index = ((T0 << 1) + (T0_frac >> 1) - (PIT_MIN<<1)); } else { index = ((T0 - PIT_FR1_8b) + ((PIT_FR1_8b - PIT_MIN)*2)); } Parm_serial(index, 8, &prms); /* find T0_min and T0_max for subframe 2 and 4 */ T0_min = (T0 - 8); if (T0_min < PIT_MIN) { T0_min = PIT_MIN; } T0_max = T0_min + 15; if (T0_max > PIT_MAX) { T0_max = PIT_MAX; T0_min = (T0_max - 15); } } else { /* if subframe 2 or 4 */ /*--------------------------------------------------------------* * The pitch range for subframe 2 or 4 is encoded with 5 bits: * * T0_min to T0_max resolution 1/2 (frac = 0 or 2) * *--------------------------------------------------------------*/ i = (T0 - T0_min); index = (i << 1) + (T0_frac >> 1); Parm_serial(index, 5, &prms); } } else { T0 = Pitch_fr4(&exc[i_subfr], xn, h1, T0_min, T0_max, &T0_frac, pit_flag, PIT_FR2, PIT_FR1_9b, L_SUBFR); /* encode pitch lag */ if (pit_flag == 0) /* if 1st/3rd subframe */ { /*--------------------------------------------------------------* * The pitch range for the 1st/3rd subframe is encoded with * * 9 bits and is divided as follows: * * PIT_MIN to PIT_FR2-1 resolution 1/4 (frac = 0,1,2 or 3) * * PIT_FR2 to PIT_FR1-1 resolution 1/2 (frac = 0 or 1) * * PIT_FR1 to PIT_MAX resolution 1 (frac = 0) * *--------------------------------------------------------------*/ if (T0 < PIT_FR2) { index = ((T0 << 2) + T0_frac) - (PIT_MIN << 2); } else if(T0 < PIT_FR1_9b) { index = ((((T0 << 1) + (T0_frac >> 1)) - (PIT_FR2<<1)) + ((PIT_FR2 - PIT_MIN)<<2)); } else { index = (((T0 - PIT_FR1_9b) + ((PIT_FR2 - PIT_MIN)<<2)) + ((PIT_FR1_9b - PIT_FR2)<<1)); } Parm_serial(index, 9, &prms); /* find T0_min and T0_max for subframe 2 and 4 */ T0_min = (T0 - 8); if (T0_min < PIT_MIN) { T0_min = PIT_MIN; } T0_max = T0_min + 15; if (T0_max > PIT_MAX) { T0_max = PIT_MAX; T0_min = (T0_max - 15); } } else { /* if subframe 2 or 4 */ /*--------------------------------------------------------------* * The pitch range for subframe 2 or 4 is encoded with 6 bits: * * T0_min to T0_max resolution 1/4 (frac = 0,1,2 or 3) * *--------------------------------------------------------------*/ i = (T0 - T0_min); index = (i << 2) + T0_frac; Parm_serial(index, 6, &prms); } } /*-----------------------------------------------------------------* * Gain clipping test to avoid unstable synthesis on frame erasure * *-----------------------------------------------------------------*/ clip_gain = 0; if((st->gp_clip[0] < 154) && (st->gp_clip[1] > 14746)) clip_gain = 1; /*-----------------------------------------------------------------* * - find unity gain pitch excitation (adaptive codebook entry) * * with fractional interpolation. * * - find filtered pitch exc. y1[]=exc[] convolved with h1[]) * * - compute pitch gain1 * *-----------------------------------------------------------------*/ /* find pitch exitation */ #ifdef ASM_OPT /* asm optimization branch */ pred_lt4_asm(&exc[i_subfr], T0, T0_frac, L_SUBFR + 1); #else Pred_lt4(&exc[i_subfr], T0, T0_frac, L_SUBFR + 1); #endif if (*ser_size > NBBITS_9k) { #ifdef ASM_OPT /* asm optimization branch */ Convolve_asm(&exc[i_subfr], h1, y1, L_SUBFR); #else Convolve(&exc[i_subfr], h1, y1, L_SUBFR); #endif gain1 = G_pitch(xn, y1, g_coeff, L_SUBFR); /* clip gain if necessary to avoid problem at decoder */ if ((clip_gain != 0) && (gain1 > GP_CLIP)) { gain1 = GP_CLIP; } /* find energy of new target xn2[] */ Updt_tar(xn, dn, y1, gain1, L_SUBFR); /* dn used temporary */ } else { gain1 = 0; } /*-----------------------------------------------------------------* * - find pitch excitation filtered by 1st order LP filter. * * - find filtered pitch exc. y2[]=exc[] convolved with h1[]) * * - compute pitch gain2 * *-----------------------------------------------------------------*/ /* find pitch excitation with lp filter */ vo_p0 = exc + i_subfr-1; vo_p1 = code; /* find pitch excitation with lp filter */ for (i = 0; i < L_SUBFR/2; i++) { L_tmp = 5898 * *vo_p0++; L_tmp1 = 5898 * *vo_p0; L_tmp += 20972 * *vo_p0++; L_tmp1 += 20972 * *vo_p0++; L_tmp1 += 5898 * *vo_p0--; L_tmp += 5898 * *vo_p0; *vo_p1++ = (L_tmp + 0x4000)>>15; *vo_p1++ = (L_tmp1 + 0x4000)>>15; } #ifdef ASM_OPT /* asm optimization branch */ Convolve_asm(code, h1, y2, L_SUBFR); #else Convolve(code, h1, y2, L_SUBFR); #endif gain2 = G_pitch(xn, y2, g_coeff2, L_SUBFR); /* clip gain if necessary to avoid problem at decoder */ if ((clip_gain != 0) && (gain2 > GP_CLIP)) { gain2 = GP_CLIP; } /* find energy of new target xn2[] */ Updt_tar(xn, xn2, y2, gain2, L_SUBFR); /*-----------------------------------------------------------------* * use the best prediction (minimise quadratic error). * *-----------------------------------------------------------------*/ select = 0; if(*ser_size > NBBITS_9k) { L_tmp = 0L; vo_p0 = dn; vo_p1 = xn2; for (i = 0; i < L_SUBFR/2; i++) { L_tmp += *vo_p0 * *vo_p0; vo_p0++; L_tmp -= *vo_p1 * *vo_p1; vo_p1++; L_tmp += *vo_p0 * *vo_p0; vo_p0++; L_tmp -= *vo_p1 * *vo_p1; vo_p1++; } if (L_tmp <= 0) { select = 1; } Parm_serial(select, 1, &prms); } if (select == 0) { /* use the lp filter for pitch excitation prediction */ gain_pit = gain2; Copy(code, &exc[i_subfr], L_SUBFR); Copy(y2, y1, L_SUBFR); Copy(g_coeff2, g_coeff, 4); } else { /* no filter used for pitch excitation prediction */ gain_pit = gain1; Copy(dn, xn2, L_SUBFR); /* target vector for codebook search */ } /*-----------------------------------------------------------------* * - update cn[] for codebook search * *-----------------------------------------------------------------*/ Updt_tar(cn, cn, &exc[i_subfr], gain_pit, L_SUBFR); #ifdef ASM_OPT /* asm optimization branch */ Scale_sig_opt(cn, L_SUBFR, shift); /* scaling of cn[] to limit dynamic at 12 bits */ #else Scale_sig(cn, L_SUBFR, shift); /* scaling of cn[] to limit dynamic at 12 bits */ #endif /*-----------------------------------------------------------------* * - include fixed-gain pitch contribution into impulse resp. h1[] * *-----------------------------------------------------------------*/ tmp = 0; Preemph(h2, st->tilt_code, L_SUBFR, &tmp); if (T0_frac > 2) T0 = (T0 + 1); Pit_shrp(h2, T0, PIT_SHARP, L_SUBFR); /*-----------------------------------------------------------------* * - Correlation between target xn2[] and impulse response h1[] * * - Innovative codebook search * *-----------------------------------------------------------------*/ cor_h_x(h2, xn2, dn); if (*ser_size <= NBBITS_7k) { ACELP_2t64_fx(dn, cn, h2, code, y2, indice); Parm_serial(indice[0], 12, &prms); } else if(*ser_size <= NBBITS_9k) { ACELP_4t64_fx(dn, cn, h2, code, y2, 20, *ser_size, indice); Parm_serial(indice[0], 5, &prms); Parm_serial(indice[1], 5, &prms); Parm_serial(indice[2], 5, &prms); Parm_serial(indice[3], 5, &prms); } else if(*ser_size <= NBBITS_12k) { ACELP_4t64_fx(dn, cn, h2, code, y2, 36, *ser_size, indice); Parm_serial(indice[0], 9, &prms); Parm_serial(indice[1], 9, &prms); Parm_serial(indice[2], 9, &prms); Parm_serial(indice[3], 9, &prms); } else if(*ser_size <= NBBITS_14k) { ACELP_4t64_fx(dn, cn, h2, code, y2, 44, *ser_size, indice); Parm_serial(indice[0], 13, &prms); Parm_serial(indice[1], 13, &prms); Parm_serial(indice[2], 9, &prms); Parm_serial(indice[3], 9, &prms); } else if(*ser_size <= NBBITS_16k) { ACELP_4t64_fx(dn, cn, h2, code, y2, 52, *ser_size, indice); Parm_serial(indice[0], 13, &prms); Parm_serial(indice[1], 13, &prms); Parm_serial(indice[2], 13, &prms); Parm_serial(indice[3], 13, &prms); } else if(*ser_size <= NBBITS_18k) { ACELP_4t64_fx(dn, cn, h2, code, y2, 64, *ser_size, indice); Parm_serial(indice[0], 2, &prms); Parm_serial(indice[1], 2, &prms); Parm_serial(indice[2], 2, &prms); Parm_serial(indice[3], 2, &prms); Parm_serial(indice[4], 14, &prms); Parm_serial(indice[5], 14, &prms); Parm_serial(indice[6], 14, &prms); Parm_serial(indice[7], 14, &prms); } else if(*ser_size <= NBBITS_20k) { ACELP_4t64_fx(dn, cn, h2, code, y2, 72, *ser_size, indice); Parm_serial(indice[0], 10, &prms); Parm_serial(indice[1], 10, &prms); Parm_serial(indice[2], 2, &prms); Parm_serial(indice[3], 2, &prms); Parm_serial(indice[4], 10, &prms); Parm_serial(indice[5], 10, &prms); Parm_serial(indice[6], 14, &prms); Parm_serial(indice[7], 14, &prms); } else { ACELP_4t64_fx(dn, cn, h2, code, y2, 88, *ser_size, indice); Parm_serial(indice[0], 11, &prms); Parm_serial(indice[1], 11, &prms); Parm_serial(indice[2], 11, &prms); Parm_serial(indice[3], 11, &prms); Parm_serial(indice[4], 11, &prms); Parm_serial(indice[5], 11, &prms); Parm_serial(indice[6], 11, &prms); Parm_serial(indice[7], 11, &prms); } /*-------------------------------------------------------* * - Add the fixed-gain pitch contribution to code[]. * *-------------------------------------------------------*/ tmp = 0; Preemph(code, st->tilt_code, L_SUBFR, &tmp); Pit_shrp(code, T0, PIT_SHARP, L_SUBFR); /*----------------------------------------------------------* * - Compute the fixed codebook gain * * - quantize fixed codebook gain * *----------------------------------------------------------*/ if(*ser_size <= NBBITS_9k) { index = Q_gain2(xn, y1, Q_new + shift, y2, code, g_coeff, L_SUBFR, 6, &gain_pit, &L_gain_code, clip_gain, st->qua_gain); Parm_serial(index, 6, &prms); } else { index = Q_gain2(xn, y1, Q_new + shift, y2, code, g_coeff, L_SUBFR, 7, &gain_pit, &L_gain_code, clip_gain, st->qua_gain); Parm_serial(index, 7, &prms); } /* test quantized gain of pitch for pitch clipping algorithm */ Gp_clip_test_gain_pit(gain_pit, st->gp_clip); L_tmp = L_shl(L_gain_code, Q_new); gain_code = extract_h(L_add(L_tmp, 0x8000)); /*----------------------------------------------------------* * Update parameters for the next subframe. * * - tilt of code: 0.0 (unvoiced) to 0.5 (voiced) * *----------------------------------------------------------*/ /* find voice factor in Q15 (1=voiced, -1=unvoiced) */ Copy(&exc[i_subfr], exc2, L_SUBFR); #ifdef ASM_OPT /* asm optimization branch */ Scale_sig_opt(exc2, L_SUBFR, shift); #else Scale_sig(exc2, L_SUBFR, shift); #endif voice_fac = voice_factor(exc2, shift, gain_pit, code, gain_code, L_SUBFR); /* tilt of code for next subframe: 0.5=voiced, 0=unvoiced */ st->tilt_code = ((voice_fac >> 2) + 8192); /*------------------------------------------------------* * - Update filter's memory "mem_w0" for finding the * * target vector in the next subframe. * * - Find the total excitation * * - Find synthesis speech to update mem_syn[]. * *------------------------------------------------------*/ /* y2 in Q9, gain_pit in Q14 */ L_tmp = (gain_code * y2[L_SUBFR - 1])<<1; L_tmp = L_shl(L_tmp, (5 + shift)); L_tmp = L_negate(L_tmp); L_tmp += (xn[L_SUBFR - 1] * 16384)<<1; L_tmp -= (y1[L_SUBFR - 1] * gain_pit)<<1; L_tmp = L_shl(L_tmp, (1 - shift)); st->mem_w0 = extract_h(L_add(L_tmp, 0x8000)); if (*ser_size >= NBBITS_24k) Copy(&exc[i_subfr], exc2, L_SUBFR); for (i = 0; i < L_SUBFR; i++) { /* code in Q9, gain_pit in Q14 */ L_tmp = (gain_code * code[i])<<1; L_tmp = (L_tmp << 5); L_tmp += (exc[i + i_subfr] * gain_pit)<<1; L_tmp = L_shl2(L_tmp, 1); exc[i + i_subfr] = extract_h(L_add(L_tmp, 0x8000)); } Syn_filt(p_Aq,&exc[i_subfr], synth, L_SUBFR, st->mem_syn, 1); if(*ser_size >= NBBITS_24k) { /*------------------------------------------------------------* * phase dispersion to enhance noise in low bit rate * *------------------------------------------------------------*/ /* L_gain_code in Q16 */ VO_L_Extract(L_gain_code, &gain_code, &gain_code_lo); /*------------------------------------------------------------* * noise enhancer * * ~~~~~~~~~~~~~~ * * - Enhance excitation on noise. (modify gain of code) * * If signal is noisy and LPC filter is stable, move gain * * of code 1.5 dB toward gain of code threshold. * * This decrease by 3 dB noise energy variation. * *------------------------------------------------------------*/ tmp = (16384 - (voice_fac >> 1)); /* 1=unvoiced, 0=voiced */ fac = vo_mult(stab_fac, tmp); L_tmp = L_gain_code; if(L_tmp < st->L_gc_thres) { L_tmp = vo_L_add(L_tmp, Mpy_32_16(gain_code, gain_code_lo, 6226)); if(L_tmp > st->L_gc_thres) { L_tmp = st->L_gc_thres; } } else { L_tmp = Mpy_32_16(gain_code, gain_code_lo, 27536); if(L_tmp < st->L_gc_thres) { L_tmp = st->L_gc_thres; } } st->L_gc_thres = L_tmp; L_gain_code = Mpy_32_16(gain_code, gain_code_lo, (32767 - fac)); VO_L_Extract(L_tmp, &gain_code, &gain_code_lo); L_gain_code = vo_L_add(L_gain_code, Mpy_32_16(gain_code, gain_code_lo, fac)); /*------------------------------------------------------------* * pitch enhancer * * ~~~~~~~~~~~~~~ * * - Enhance excitation on voice. (HP filtering of code) * * On voiced signal, filtering of code by a smooth fir HP * * filter to decrease energy of code in low frequency. * *------------------------------------------------------------*/ tmp = ((voice_fac >> 3) + 4096); /* 0.25=voiced, 0=unvoiced */ L_tmp = L_deposit_h(code[0]); L_tmp -= (code[1] * tmp)<<1; code2[0] = vo_round(L_tmp); for (i = 1; i < L_SUBFR - 1; i++) { L_tmp = L_deposit_h(code[i]); L_tmp -= (code[i + 1] * tmp)<<1; L_tmp -= (code[i - 1] * tmp)<<1; code2[i] = vo_round(L_tmp); } L_tmp = L_deposit_h(code[L_SUBFR - 1]); L_tmp -= (code[L_SUBFR - 2] * tmp)<<1; code2[L_SUBFR - 1] = vo_round(L_tmp); /* build excitation */ gain_code = vo_round(L_shl(L_gain_code, Q_new)); for (i = 0; i < L_SUBFR; i++) { L_tmp = (code2[i] * gain_code)<<1; L_tmp = (L_tmp << 5); L_tmp += (exc2[i] * gain_pit)<<1; L_tmp = (L_tmp << 1); exc2[i] = vo_round(L_tmp); } corr_gain = synthesis(p_Aq, exc2, Q_new, &speech16k[i_subfr * 5 / 4], st); Parm_serial(corr_gain, 4, &prms); } p_A += (M + 1); p_Aq += (M + 1); } /* end of subframe loop */ /*--------------------------------------------------* * Update signal for next frame. * * -> save past of speech[], wsp[] and exc[]. * *--------------------------------------------------*/ Copy(&old_speech[L_FRAME], st->old_speech, L_TOTAL - L_FRAME); Copy(&old_wsp[L_FRAME / OPL_DECIM], st->old_wsp, PIT_MAX / OPL_DECIM); Copy(&old_exc[L_FRAME], st->old_exc, PIT_MAX + L_INTERPOL); return; } /*-----------------------------------------------------* * Function synthesis() * * * * Synthesis of signal at 16kHz with HF extension. * * * *-----------------------------------------------------*/ static Word16 synthesis( Word16 Aq[], /* A(z) : quantized Az */ Word16 exc[], /* (i) : excitation at 12kHz */ Word16 Q_new, /* (i) : scaling performed on exc */ Word16 synth16k[], /* (o) : 16kHz synthesis signal */ Coder_State * st /* (i/o) : State structure */ ) { Word16 fac, tmp, exp; Word16 ener, exp_ener; Word32 L_tmp, i; Word16 synth_hi[M + L_SUBFR], synth_lo[M + L_SUBFR]; Word16 synth[L_SUBFR]; Word16 HF[L_SUBFR16k]; /* High Frequency vector */ Word16 Ap[M + 1]; Word16 HF_SP[L_SUBFR16k]; /* High Frequency vector (from original signal) */ Word16 HP_est_gain, HP_calc_gain, HP_corr_gain; Word16 dist_min, dist; Word16 HP_gain_ind = 0; Word16 gain1, gain2; Word16 weight1, weight2; /*------------------------------------------------------------* * speech synthesis * * ~~~~~~~~~~~~~~~~ * * - Find synthesis speech corresponding to exc2[]. * * - Perform fixed deemphasis and hp 50hz filtering. * * - Oversampling from 12.8kHz to 16kHz. * *------------------------------------------------------------*/ Copy(st->mem_syn_hi, synth_hi, M); Copy(st->mem_syn_lo, synth_lo, M); #ifdef ASM_OPT /* asm optimization branch */ Syn_filt_32_asm(Aq, M, exc, Q_new, synth_hi + M, synth_lo + M, L_SUBFR); #else Syn_filt_32(Aq, M, exc, Q_new, synth_hi + M, synth_lo + M, L_SUBFR); #endif Copy(synth_hi + L_SUBFR, st->mem_syn_hi, M); Copy(synth_lo + L_SUBFR, st->mem_syn_lo, M); #ifdef ASM_OPT /* asm optimization branch */ Deemph_32_asm(synth_hi + M, synth_lo + M, synth, &(st->mem_deemph)); #else Deemph_32(synth_hi + M, synth_lo + M, synth, PREEMPH_FAC, L_SUBFR, &(st->mem_deemph)); #endif HP50_12k8(synth, L_SUBFR, st->mem_sig_out); /* Original speech signal as reference for high band gain quantisation */ for (i = 0; i < L_SUBFR16k; i++) { HF_SP[i] = synth16k[i]; } /*------------------------------------------------------* * HF noise synthesis * * ~~~~~~~~~~~~~~~~~~ * * - Generate HF noise between 5.5 and 7.5 kHz. * * - Set energy of noise according to synthesis tilt. * * tilt > 0.8 ==> - 14 dB (voiced) * * tilt 0.5 ==> - 6 dB (voiced or noise) * * tilt < 0.0 ==> 0 dB (noise) * *------------------------------------------------------*/ /* generate white noise vector */ for (i = 0; i < L_SUBFR16k; i++) { HF[i] = Random(&(st->seed2))>>3; } /* energy of excitation */ #ifdef ASM_OPT /* asm optimization branch */ Scale_sig_opt(exc, L_SUBFR, -3); Q_new = Q_new - 3; ener = extract_h(Dot_product12_asm(exc, exc, L_SUBFR, &exp_ener)); #else Scale_sig(exc, L_SUBFR, -3); Q_new = Q_new - 3; ener = extract_h(Dot_product12(exc, exc, L_SUBFR, &exp_ener)); #endif exp_ener = exp_ener - (Q_new + Q_new); /* set energy of white noise to energy of excitation */ #ifdef ASM_OPT /* asm optimization branch */ tmp = extract_h(Dot_product12_asm(HF, HF, L_SUBFR16k, &exp)); #else tmp = extract_h(Dot_product12(HF, HF, L_SUBFR16k, &exp)); #endif if(tmp > ener) { tmp = (tmp >> 1); /* Be sure tmp < ener */ exp = (exp + 1); } L_tmp = L_deposit_h(div_s(tmp, ener)); /* result is normalized */ exp = (exp - exp_ener); Isqrt_n(&L_tmp, &exp); L_tmp = L_shl(L_tmp, (exp + 1)); /* L_tmp x 2, L_tmp in Q31 */ tmp = extract_h(L_tmp); /* tmp = 2 x sqrt(ener_exc/ener_hf) */ for (i = 0; i < L_SUBFR16k; i++) { HF[i] = vo_mult(HF[i], tmp); } /* find tilt of synthesis speech (tilt: 1=voiced, -1=unvoiced) */ HP400_12k8(synth, L_SUBFR, st->mem_hp400); L_tmp = 1L; for (i = 0; i < L_SUBFR; i++) L_tmp += (synth[i] * synth[i])<<1; exp = norm_l(L_tmp); ener = extract_h(L_tmp << exp); /* ener = r[0] */ L_tmp = 1L; for (i = 1; i < L_SUBFR; i++) L_tmp +=(synth[i] * synth[i - 1])<<1; tmp = extract_h(L_tmp << exp); /* tmp = r[1] */ if (tmp > 0) { fac = div_s(tmp, ener); } else { fac = 0; } /* modify energy of white noise according to synthesis tilt */ gain1 = 32767 - fac; gain2 = vo_mult(gain1, 20480); gain2 = shl(gain2, 1); if (st->vad_hist > 0) { weight1 = 0; weight2 = 32767; } else { weight1 = 32767; weight2 = 0; } tmp = vo_mult(weight1, gain1); tmp = add1(tmp, vo_mult(weight2, gain2)); if (tmp != 0) { tmp = (tmp + 1); } HP_est_gain = tmp; if(HP_est_gain < 3277) { HP_est_gain = 3277; /* 0.1 in Q15 */ } /* synthesis of noise: 4.8kHz..5.6kHz --> 6kHz..7kHz */ Weight_a(Aq, Ap, 19661, M); /* fac=0.6 */ #ifdef ASM_OPT /* asm optimization branch */ Syn_filt_asm(Ap, HF, HF, st->mem_syn_hf); /* noise High Pass filtering (1ms of delay) */ Filt_6k_7k_asm(HF, L_SUBFR16k, st->mem_hf); /* filtering of the original signal */ Filt_6k_7k_asm(HF_SP, L_SUBFR16k, st->mem_hf2); /* check the gain difference */ Scale_sig_opt(HF_SP, L_SUBFR16k, -1); ener = extract_h(Dot_product12_asm(HF_SP, HF_SP, L_SUBFR16k, &exp_ener)); /* set energy of white noise to energy of excitation */ tmp = extract_h(Dot_product12_asm(HF, HF, L_SUBFR16k, &exp)); #else Syn_filt(Ap, HF, HF, L_SUBFR16k, st->mem_syn_hf, 1); /* noise High Pass filtering (1ms of delay) */ Filt_6k_7k(HF, L_SUBFR16k, st->mem_hf); /* filtering of the original signal */ Filt_6k_7k(HF_SP, L_SUBFR16k, st->mem_hf2); /* check the gain difference */ Scale_sig(HF_SP, L_SUBFR16k, -1); ener = extract_h(Dot_product12(HF_SP, HF_SP, L_SUBFR16k, &exp_ener)); /* set energy of white noise to energy of excitation */ tmp = extract_h(Dot_product12(HF, HF, L_SUBFR16k, &exp)); #endif if (tmp > ener) { tmp = (tmp >> 1); /* Be sure tmp < ener */ exp = (exp + 1); } L_tmp = L_deposit_h(div_s(tmp, ener)); /* result is normalized */ exp = vo_sub(exp, exp_ener); Isqrt_n(&L_tmp, &exp); L_tmp = L_shl(L_tmp, exp); /* L_tmp, L_tmp in Q31 */ HP_calc_gain = extract_h(L_tmp); /* tmp = sqrt(ener_input/ener_hf) */ /* st->gain_alpha *= st->dtx_encSt->dtxHangoverCount/7 */ L_tmp = (vo_L_mult(st->dtx_encSt->dtxHangoverCount, 4681) << 15); st->gain_alpha = vo_mult(st->gain_alpha, extract_h(L_tmp)); if(st->dtx_encSt->dtxHangoverCount > 6) st->gain_alpha = 32767; HP_est_gain = HP_est_gain >> 1; /* From Q15 to Q14 */ HP_corr_gain = add1(vo_mult(HP_calc_gain, st->gain_alpha), vo_mult((32767 - st->gain_alpha), HP_est_gain)); /* Quantise the correction gain */ dist_min = 32767; for (i = 0; i < 16; i++) { dist = vo_mult((HP_corr_gain - HP_gain[i]), (HP_corr_gain - HP_gain[i])); if (dist_min > dist) { dist_min = dist; HP_gain_ind = i; } } HP_corr_gain = HP_gain[HP_gain_ind]; /* return the quantised gain index when using the highest mode, otherwise zero */ return (HP_gain_ind); } /************************************************* * * Breif: Codec main function * **************************************************/ int AMR_Enc_Encode(HAMRENC hCodec) { Word32 i; Coder_State *gData = (Coder_State*)hCodec; Word16 *signal; Word16 packed_size = 0; Word16 prms[NB_BITS_MAX]; Word16 coding_mode = 0, nb_bits, allow_dtx, mode, reset_flag; mode = gData->mode; coding_mode = gData->mode; nb_bits = nb_of_bits[mode]; signal = (Word16 *)gData->inputStream; allow_dtx = gData->allow_dtx; /* check for homing frame */ reset_flag = encoder_homing_frame_test(signal); for (i = 0; i < L_FRAME16k; i++) /* Delete the 2 LSBs (14-bit input) */ { *(signal + i) = (Word16) (*(signal + i) & 0xfffC); } coder(&coding_mode, signal, prms, &nb_bits, gData, allow_dtx); packed_size = PackBits(prms, coding_mode, mode, gData); if (reset_flag != 0) { Reset_encoder(gData, 1); } return packed_size; } /*************************************************************************** * *Brief: Codec API function --- Initialize the codec and return a codec handle * ***************************************************************************/ VO_U32 VO_API voAMRWB_Init(VO_HANDLE * phCodec, /* o: the audio codec handle */ VO_AUDIO_CODINGTYPE vType, /* i: Codec Type ID */ VO_CODEC_INIT_USERDATA * pUserData /* i: init Parameters */ ) { Coder_State *st; FrameStream *stream; #ifdef USE_DEAULT_MEM VO_MEM_OPERATOR voMemoprator; #endif VO_MEM_OPERATOR *pMemOP; int interMem = 0; if(pUserData == NULL || pUserData->memflag != VO_IMF_USERMEMOPERATOR || pUserData->memData == NULL ) { #ifdef USE_DEAULT_MEM voMemoprator.Alloc = cmnMemAlloc; voMemoprator.Copy = cmnMemCopy; voMemoprator.Free = cmnMemFree; voMemoprator.Set = cmnMemSet; voMemoprator.Check = cmnMemCheck; interMem = 1; pMemOP = &voMemoprator; #else *phCodec = NULL; return VO_ERR_INVALID_ARG; #endif } else { pMemOP = (VO_MEM_OPERATOR *)pUserData->memData; } /*-------------------------------------------------------------------------* * Memory allocation for coder state. * *-------------------------------------------------------------------------*/ if ((st = (Coder_State *)mem_malloc(pMemOP, sizeof(Coder_State), 32, VO_INDEX_ENC_AMRWB)) == NULL) { return VO_ERR_OUTOF_MEMORY; } st->vadSt = NULL; st->dtx_encSt = NULL; st->sid_update_counter = 3; st->sid_handover_debt = 0; st->prev_ft = TX_SPEECH; st->inputStream = NULL; st->inputSize = 0; /* Default setting */ st->mode = VOAMRWB_MD2385; /* bit rate 23.85kbps */ st->frameType = VOAMRWB_RFC3267; /* frame type: RFC3267 */ st->allow_dtx = 0; /* disable DTX mode */ st->outputStream = NULL; st->outputSize = 0; st->stream = (FrameStream *)mem_malloc(pMemOP, sizeof(FrameStream), 32, VO_INDEX_ENC_AMRWB); if(st->stream == NULL) return VO_ERR_OUTOF_MEMORY; st->stream->frame_ptr = (unsigned char *)mem_malloc(pMemOP, Frame_Maxsize, 32, VO_INDEX_ENC_AMRWB); if(st->stream->frame_ptr == NULL) return VO_ERR_OUTOF_MEMORY; stream = st->stream; voAWB_InitFrameBuffer(stream); wb_vad_init(&(st->vadSt), pMemOP); dtx_enc_init(&(st->dtx_encSt), isf_init, pMemOP); Reset_encoder((void *) st, 1); if(interMem) { st->voMemoprator.Alloc = cmnMemAlloc; st->voMemoprator.Copy = cmnMemCopy; st->voMemoprator.Free = cmnMemFree; st->voMemoprator.Set = cmnMemSet; st->voMemoprator.Check = cmnMemCheck; pMemOP = &st->voMemoprator; } st->pvoMemop = pMemOP; *phCodec = (void *) st; return VO_ERR_NONE; } /********************************************************************************** * * Brief: Codec API function: Input PCM data * ***********************************************************************************/ VO_U32 VO_API voAMRWB_SetInputData( VO_HANDLE hCodec, /* i/o: The codec handle which was created by Init function */ VO_CODECBUFFER * pInput /* i: The input buffer parameter */ ) { Coder_State *gData; FrameStream *stream; if(NULL == hCodec) { return VO_ERR_INVALID_ARG; } gData = (Coder_State *)hCodec; stream = gData->stream; if(NULL == pInput || NULL == pInput->Buffer) { return VO_ERR_INVALID_ARG; } stream->set_ptr = pInput->Buffer; stream->set_len = pInput->Length; stream->frame_ptr = stream->frame_ptr_bk; stream->used_len = 0; return VO_ERR_NONE; } /************************************************************************************** * * Brief: Codec API function: Get the compression audio data frame by frame * ***************************************************************************************/ VO_U32 VO_API voAMRWB_GetOutputData( VO_HANDLE hCodec, /* i: The Codec Handle which was created by Init function*/ VO_CODECBUFFER * pOutput, /* o: The output audio data */ VO_AUDIO_OUTPUTINFO * pAudioFormat /* o: The encoder module filled audio format and used the input size*/ ) { Coder_State* gData = (Coder_State*)hCodec; VO_MEM_OPERATOR *pMemOP; FrameStream *stream = (FrameStream *)gData->stream; pMemOP = (VO_MEM_OPERATOR *)gData->pvoMemop; if(stream->framebuffer_len < Frame_MaxByte) /* check the work buffer len */ { stream->frame_storelen = stream->framebuffer_len; if(stream->frame_storelen) { pMemOP->Copy(VO_INDEX_ENC_AMRWB, stream->frame_ptr_bk , stream->frame_ptr , stream->frame_storelen); } if(stream->set_len > 0) { voAWB_UpdateFrameBuffer(stream, pMemOP); } if(stream->framebuffer_len < Frame_MaxByte) { if(pAudioFormat) pAudioFormat->InputUsed = stream->used_len; return VO_ERR_INPUT_BUFFER_SMALL; } } gData->inputStream = stream->frame_ptr; gData->outputStream = (unsigned short*)pOutput->Buffer; gData->outputSize = AMR_Enc_Encode(gData); /* encoder main function */ pOutput->Length = gData->outputSize; /* get the output buffer length */ stream->frame_ptr += 640; /* update the work buffer ptr */ stream->framebuffer_len -= 640; if(pAudioFormat) /* return output audio information */ { pAudioFormat->Format.Channels = 1; pAudioFormat->Format.SampleRate = 8000; pAudioFormat->Format.SampleBits = 16; pAudioFormat->InputUsed = stream->used_len; } return VO_ERR_NONE; } /************************************************************************* * * Brief: Codec API function---set the data by specified parameter ID * *************************************************************************/ VO_U32 VO_API voAMRWB_SetParam( VO_HANDLE hCodec, /* i/o: The Codec Handle which was created by Init function */ VO_S32 uParamID, /* i: The param ID */ VO_PTR pData /* i: The param value depend on the ID */ ) { Coder_State* gData = (Coder_State*)hCodec; FrameStream *stream = (FrameStream *)(gData->stream); int *lValue = (int*)pData; switch(uParamID) { /* setting AMR-WB frame type*/ case VO_PID_AMRWB_FRAMETYPE: if(*lValue < VOAMRWB_DEFAULT || *lValue > VOAMRWB_RFC3267) return VO_ERR_WRONG_PARAM_ID; gData->frameType = *lValue; break; /* setting AMR-WB bit rate */ case VO_PID_AMRWB_MODE: { if(*lValue < VOAMRWB_MD66 || *lValue > VOAMRWB_MD2385) return VO_ERR_WRONG_PARAM_ID; gData->mode = *lValue; } break; /* enable or disable DTX mode */ case VO_PID_AMRWB_DTX: gData->allow_dtx = (Word16)(*lValue); break; case VO_PID_COMMON_HEADDATA: break; /* flush the work buffer */ case VO_PID_COMMON_FLUSH: stream->set_ptr = NULL; stream->frame_storelen = 0; stream->framebuffer_len = 0; stream->set_len = 0; break; default: return VO_ERR_WRONG_PARAM_ID; } return VO_ERR_NONE; } /************************************************************************** * *Brief: Codec API function---Get the data by specified parameter ID * ***************************************************************************/ VO_U32 VO_API voAMRWB_GetParam( VO_HANDLE hCodec, /* i: The Codec Handle which was created by Init function */ VO_S32 uParamID, /* i: The param ID */ VO_PTR pData /* o: The param value depend on the ID */ ) { int temp; Coder_State* gData = (Coder_State*)hCodec; if (gData==NULL) return VO_ERR_INVALID_ARG; switch(uParamID) { /* output audio format */ case VO_PID_AMRWB_FORMAT: { VO_AUDIO_FORMAT* fmt = (VO_AUDIO_FORMAT*)pData; fmt->Channels = 1; fmt->SampleRate = 16000; fmt->SampleBits = 16; break; } /* output audio channel number */ case VO_PID_AMRWB_CHANNELS: temp = 1; pData = (void *)(&temp); break; /* output audio sample rate */ case VO_PID_AMRWB_SAMPLERATE: temp = 16000; pData = (void *)(&temp); break; /* output audio frame type */ case VO_PID_AMRWB_FRAMETYPE: temp = gData->frameType; pData = (void *)(&temp); break; /* output audio bit rate */ case VO_PID_AMRWB_MODE: temp = gData->mode; pData = (void *)(&temp); break; default: return VO_ERR_WRONG_PARAM_ID; } return VO_ERR_NONE; } /*********************************************************************************** * * Brief: Codec API function---Release the codec after all encoder operations are done * *************************************************************************************/ VO_U32 VO_API voAMRWB_Uninit(VO_HANDLE hCodec /* i/o: Codec handle pointer */ ) { Coder_State* gData = (Coder_State*)hCodec; VO_MEM_OPERATOR *pMemOP; pMemOP = gData->pvoMemop; if(hCodec) { if(gData->stream) { if(gData->stream->frame_ptr_bk) { mem_free(pMemOP, gData->stream->frame_ptr_bk, VO_INDEX_ENC_AMRWB); gData->stream->frame_ptr_bk = NULL; } mem_free(pMemOP, gData->stream, VO_INDEX_ENC_AMRWB); gData->stream = NULL; } wb_vad_exit(&(((Coder_State *) gData)->vadSt), pMemOP); dtx_enc_exit(&(((Coder_State *) gData)->dtx_encSt), pMemOP); mem_free(pMemOP, hCodec, VO_INDEX_ENC_AMRWB); hCodec = NULL; } return VO_ERR_NONE; } /******************************************************************************** * * Brief: voGetAMRWBEncAPI gets the API handle of the codec * ********************************************************************************/ VO_S32 VO_API voGetAMRWBEncAPI( VO_AUDIO_CODECAPI * pEncHandle /* i/o: Codec handle pointer */ ) { if(NULL == pEncHandle) return VO_ERR_INVALID_ARG; pEncHandle->Init = voAMRWB_Init; pEncHandle->SetInputData = voAMRWB_SetInputData; pEncHandle->GetOutputData = voAMRWB_GetOutputData; pEncHandle->SetParam = voAMRWB_SetParam; pEncHandle->GetParam = voAMRWB_GetParam; pEncHandle->Uninit = voAMRWB_Uninit; return VO_ERR_NONE; } #ifdef __cplusplus } #endif