1 /* Copyright (c) 2023 Amazon */ 2 /* 3 Redistribution and use in source and binary forms, with or without 4 modification, are permitted provided that the following conditions 5 are met: 6 7 - Redistributions of source code must retain the above copyright 8 notice, this list of conditions and the following disclaimer. 9 10 - Redistributions in binary form must reproduce the above copyright 11 notice, this list of conditions and the following disclaimer in the 12 documentation and/or other materials provided with the distribution. 13 14 THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS 15 ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT 16 LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR 17 A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER 18 OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, 19 EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, 20 PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR 21 PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF 22 LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING 23 NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS 24 SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. 25 */ 26 27 #ifndef FWGAN_H 28 #define FWGAN_H 29 30 #include "freq.h" 31 #include "fwgan_data.h" 32 33 #define FWGAN_CONT_SAMPLES 320 34 #define NB_SUBFRAMES 4 35 #define SUBFRAME_SIZE 40 36 #define FWGAN_FRAME_SIZE (NB_SUBFRAMES*SUBFRAME_SIZE) 37 #define CONT_PCM_INPUTS 320 38 #define MAX_CONT_SIZE CONT_NET_0_OUT_SIZE 39 #define FWGAN_GAMMA 0.92f 40 #define FWGAN_DEEMPHASIS 0.85f 41 42 /* FIXME: Derive those from the model rather than hardcoding. */ 43 #define FWC1_STATE_SIZE 512 44 #define FWC2_STATE_SIZE 512 45 #define FWC3_STATE_SIZE 256 46 #define FWC4_STATE_SIZE 256 47 #define FWC5_STATE_SIZE 128 48 #define FWC6_STATE_SIZE 128 49 #define FWC7_STATE_SIZE 80 50 51 typedef struct { 52 FWGAN model; 53 int arch; 54 int cont_initialized; 55 float embed_phase[2]; 56 float last_gain; 57 float last_lpc[LPC_ORDER]; 58 float syn_mem[LPC_ORDER]; 59 float preemph_mem; 60 float deemph_mem; 61 float pcm_buf[FWGAN_FRAME_SIZE]; 62 float cont[CONT_NET_10_OUT_SIZE]; 63 float cont_conv1_mem[FEAT_IN_CONV1_CONV_STATE_SIZE]; 64 float rnn_state[RNN_GRU_STATE_SIZE]; 65 float fwc1_state[FWC1_STATE_SIZE]; 66 float fwc2_state[FWC2_STATE_SIZE]; 67 float fwc3_state[FWC3_STATE_SIZE]; 68 float fwc4_state[FWC4_STATE_SIZE]; 69 float fwc5_state[FWC5_STATE_SIZE]; 70 float fwc6_state[FWC6_STATE_SIZE]; 71 float fwc7_state[FWC7_STATE_SIZE]; 72 } FWGANState; 73 74 void fwgan_init(FWGANState *st); 75 int fwgan_load_model(FWGANState *st, const unsigned char *data, int len); 76 77 void fwgan_cont(FWGANState *st, const float *pcm0, const float *features0); 78 79 void fwgan_synthesize(FWGANState *st, float *pcm, const float *features); 80 void fwgan_synthesize_int(FWGANState *st, opus_int16 *pcm, const float *features); 81 82 83 #endif /* FWGAN_H */ 84