/* * LSP computing for ACELP-based codecs * * Copyright (c) 2008 Vladimir Voroshilov * * This file is part of FFmpeg. * * FFmpeg is free software; you can redistribute it and/or * modify it under the terms of the GNU Lesser General Public * License as published by the Free Software Foundation; either * version 2.1 of the License, or (at your option) any later version. * * FFmpeg is distributed in the hope that it will be useful, * but WITHOUT ANY WARRANTY; without even the implied warranty of * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU * Lesser General Public License for more details. * * You should have received a copy of the GNU Lesser General Public * License along with FFmpeg; if not, write to the Free Software * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef AVCODEC_LSP_H #define AVCODEC_LSP_H #include <stdint.h> /** (I.F) means fixed-point value with F fractional and I integer bits */ /** * \brief ensure a minimum distance between LSFs * \param[in,out] lsfq LSF to check and adjust * \param lsfq_min_distance minimum distance between LSFs * \param lsfq_min minimum allowed LSF value * \param lsfq_max maximum allowed LSF value * \param lp_order LP filter order */ void ff_acelp_reorder_lsf(int16_t* lsfq, int lsfq_min_distance, int lsfq_min, int lsfq_max, int lp_order); /** * Adjust the quantized LSFs so they are increasing and not too close. * * This step is not mentioned in the AMR spec but is in the reference C decoder. * Omitting this step creates audible distortion on the sinusoidal sweep * test vectors in 3GPP TS 26.074. * * @param[in,out] lsf LSFs in Hertz * @param min_spacing minimum distance between two consecutive lsf values * @param size size of the lsf vector */ void ff_set_min_dist_lsf(float *lsf, double min_spacing, int size); /** * \brief Convert LSF to LSP * \param[out] lsp LSP coefficients (-0x8000 <= (0.15) < 0x8000) * \param lsf normalized LSF coefficients (0 <= (2.13) < 0x2000 * PI) * \param lp_order LP filter order * * \remark It is safe to pass the same array into the lsf and lsp parameters. */ void ff_acelp_lsf2lsp(int16_t *lsp, const int16_t *lsf, int lp_order); /** * Floating point version of ff_acelp_lsf2lsp() */ void ff_acelp_lsf2lspd(double *lsp, const float *lsf, int lp_order); /** * \brief LSP to LP conversion (3.2.6 of G.729) * \param[out] lp decoded LP coefficients (-0x8000 <= (3.12) < 0x8000) * \param lsp LSP coefficients (-0x8000 <= (0.15) < 0x8000) * \param lp_half_order LP filter order, divided by 2 */ void ff_acelp_lsp2lpc(int16_t* lp, const int16_t* lsp, int lp_half_order); /** * LSP to LP conversion (5.2.4 of AMR-WB) */ void ff_amrwb_lsp2lpc(const double *lsp, float *lp, int lp_order); /** * \brief Interpolate LSP for the first subframe and convert LSP -> LP for both subframes (3.2.5 and 3.2.6 of G.729) * \param[out] lp_1st decoded LP coefficients for first subframe (-0x8000 <= (3.12) < 0x8000) * \param[out] lp_2nd decoded LP coefficients for second subframe (-0x8000 <= (3.12) < 0x8000) * \param lsp_2nd LSP coefficients of the second subframe (-0x8000 <= (0.15) < 0x8000) * \param lsp_prev LSP coefficients from the second subframe of the previous frame (-0x8000 <= (0.15) < 0x8000) * \param lp_order LP filter order */ void ff_acelp_lp_decode(int16_t* lp_1st, int16_t* lp_2nd, const int16_t* lsp_2nd, const int16_t* lsp_prev, int lp_order); #define MAX_LP_HALF_ORDER 8 #define MAX_LP_ORDER (2*MAX_LP_HALF_ORDER) /** * Reconstruct LPC coefficients from the line spectral pair frequencies. * * @param lsp line spectral pairs in cosine domain * @param lpc linear predictive coding coefficients * @param lp_half_order half the number of the amount of LPCs to be * reconstructed, need to be smaller or equal to MAX_LP_HALF_ORDER * * @note buffers should have a minimux size of 2*lp_half_order elements. * * TIA/EIA/IS-733 2.4.3.3.5 */ void ff_acelp_lspd2lpc(const double *lsp, float *lpc, int lp_half_order); /** * Sort values in ascending order. * * @note O(n) if data already sorted, O(n^2) - otherwise */ void ff_sort_nearly_sorted_floats(float *vals, int len); /** * Compute the Pa / (1 + z(-1)) or Qa / (1 - z(-1)) coefficients * needed for LSP to LPC conversion. * We only need to calculate the 6 first elements of the polynomial. * * @param lsp line spectral pairs in cosine domain * @param[out] f polynomial input/output as a vector * * TIA/EIA/IS-733 2.4.3.3.5-1/2 */ void ff_lsp2polyf(const double *lsp, double *f, int lp_half_order); #endif /* AVCODEC_LSP_H */