X-Git-Url: https://git.sesse.net/?p=c64tapwav;a=blobdiff_plain;f=decode.cpp;h=ef8630de9948532f66e324d82a35485f74e24aa9;hp=80921b29c9fc9d1a1c0e742bf97f1c9e3b360540;hb=6d341e3e890484f23ce2b4ab057f344f13e32811;hpb=fd62bd13eca89fafd42bcdf9395b9a85aa7d82b1 diff --git a/decode.cpp b/decode.cpp index 80921b2..ef8630d 100644 --- a/decode.cpp +++ b/decode.cpp @@ -1,3 +1,6 @@ +// Copyright Steinar H. Gunderson +// Licensed under the GPL, v2. (See the file COPYING.) + #include #include #include @@ -9,7 +12,9 @@ #include "audioreader.h" #include "interpolate.h" +#include "level.h" #include "tap.h" +#include "filter.h" #define BUFSIZE 4096 #define C64_FREQUENCY 985248 @@ -18,9 +23,38 @@ #define SYNC_PULSE_LENGTH 378.0 #define SYNC_TEST_TOLERANCE 1.10 +// SPSA options +#define NUM_FILTER_COEFF 32 +#define NUM_ITER 5000 +#define A NUM_ITER/10 // approx +#define INITIAL_A 0.005 // A bit of trial and error... +#define INITIAL_C 0.02 // This too. +#define GAMMA 0.166 +#define ALPHA 1.0 + static float hysteresis_limit = 3000.0 / 32768.0; static bool do_calibrate = true; static bool output_cycles_plot = false; +static bool do_crop = false; +static float crop_start = 0.0f, crop_end = HUGE_VAL; + +static bool use_fir_filter = false; +static float filter_coeff[NUM_FILTER_COEFF] = { 1.0f }; // The rest is filled with 0. +static bool use_rc_filter = false; +static float rc_filter_freq; +static bool output_filtered = false; + +static bool quiet = false; +static bool do_auto_level = false; +static bool output_leveled = false; +static std::vector train_snap_points; +static bool do_train = false; + +// The minimum estimated sound level (for do_auto_level) at any given point. +// If you decrease this, you'll be able to amplify really silent signals +// by more, but you'll also increase the level of silent (ie. noise-only) segments, +// possibly caused misdetected pulses in these segments. +static float min_level = 0.05f; // between [x,x+1] double find_zerocrossing(const std::vector &pcm, int x) @@ -83,8 +117,10 @@ double calibrate(const std::vector &pulses) { sync_pulse_end = try_end; sync_pulse_stddev = stddev; } - fprintf(stderr, "Sync pulse length standard deviation: %.2f cycles\n", - sync_pulse_stddev); + if (!quiet) { + fprintf(stderr, "Sync pulse length standard deviation: %.2f cycles\n", + sync_pulse_stddev); + } double sum = 0.0; for (int i = SYNC_PULSE_START; i < sync_pulse_end; ++i) { @@ -92,8 +128,10 @@ double calibrate(const std::vector &pulses) { } double mean_length = C64_FREQUENCY * sum / (sync_pulse_end - SYNC_PULSE_START); double calibration_factor = SYNC_PULSE_LENGTH / mean_length; - fprintf(stderr, "Calibrated sync pulse length: %.2f -> %.2f (change %+.2f%%)\n", - mean_length, SYNC_PULSE_LENGTH, 100.0 * (calibration_factor - 1.0)); + if (!quiet) { + fprintf(stderr, "Calibrated sync pulse length: %.2f -> %.2f (change %+.2f%%)\n", + mean_length, SYNC_PULSE_LENGTH, 100.0 * (calibration_factor - 1.0)); + } // Check for pulses outside +/- 10% (sign of misdetection). for (int i = SYNC_PULSE_START; i < sync_pulse_end; ++i) { @@ -139,9 +177,16 @@ void output_tap(const std::vector& pulses, double calibration_factor) } static struct option long_options[] = { + {"auto-level", 0, 0, 'a' }, + {"output-leveled", 0, 0, 'A' }, {"no-calibrate", 0, 0, 's' }, {"plot-cycles", 0, 0, 'p' }, {"hysteresis-limit", required_argument, 0, 'l' }, + {"filter", required_argument, 0, 'f' }, + {"rc-filter", required_argument, 0, 'r' }, + {"output-filtered", 0, 0, 'F' }, + {"crop", required_argument, 0, 'c' }, + {"quiet", 0, 0, 'q' }, {"help", 0, 0, 'h' }, {0, 0, 0, 0 } }; @@ -150,9 +195,19 @@ void help() { fprintf(stderr, "decode [OPTIONS] AUDIO-FILE > TAP-FILE\n"); fprintf(stderr, "\n"); + fprintf(stderr, " -a, --auto-level automatically adjust amplitude levels throughout the file\n"); + fprintf(stderr, " -A, --output-leveled output leveled waveform to leveled.raw\n"); + fprintf(stderr, " -m, --min-level minimum estimated sound level (0..32768) for --auto-level\n"); fprintf(stderr, " -s, --no-calibrate do not try to calibrate on sync pulse length\n"); fprintf(stderr, " -p, --plot-cycles output debugging info to cycles.plot\n"); fprintf(stderr, " -l, --hysteresis-limit VAL change amplitude threshold for ignoring pulses (0..32768)\n"); + fprintf(stderr, " -f, --filter C1:C2:C3:... specify FIR filter (up to %d coefficients)\n", NUM_FILTER_COEFF); + fprintf(stderr, " -r, --rc-filter FREQ send signal through a highpass RC filter with given frequency (in Hertz)\n"); + fprintf(stderr, " -F, --output-filtered output filtered waveform to filtered.raw\n"); + fprintf(stderr, " -c, --crop START[:END] use only the given part of the file\n"); + fprintf(stderr, " -t, --train LEN1:LEN2:... train a filter for detecting any of the given number of cycles\n"); + fprintf(stderr, " (implies --no-calibrate and --quiet unless overridden)\n"); + fprintf(stderr, " -q, --quiet suppress some informational messages\n"); fprintf(stderr, " -h, --help display this help, then exit\n"); exit(1); } @@ -161,11 +216,23 @@ void parse_options(int argc, char **argv) { for ( ;; ) { int option_index = 0; - int c = getopt_long(argc, argv, "spl:h", long_options, &option_index); + int c = getopt_long(argc, argv, "aAm:spl:f:r:Fc:t:qh", long_options, &option_index); if (c == -1) break; switch (c) { + case 'a': + do_auto_level = true; + break; + + case 'A': + output_leveled = true; + break; + + case 'm': + min_level = atof(optarg) / 32768.0; + break; + case 's': do_calibrate = false; break; @@ -173,10 +240,62 @@ void parse_options(int argc, char **argv) case 'p': output_cycles_plot = true; break; + case 'l': hysteresis_limit = atof(optarg) / 32768.0; break; + case 'f': { + const char *coeffstr = strtok(optarg, ": "); + int coeff_index = 0; + while (coeff_index < NUM_FILTER_COEFF && coeffstr != NULL) { + filter_coeff[coeff_index++] = atof(coeffstr); + coeffstr = strtok(NULL, ": "); + } + use_fir_filter = true; + break; + } + + case 'r': + use_rc_filter = true; + rc_filter_freq = atof(optarg); + break; + + case 'F': + output_filtered = true; + break; + + case 'c': { + const char *cropstr = strtok(optarg, ":"); + crop_start = atof(cropstr); + cropstr = strtok(NULL, ":"); + if (cropstr == NULL) { + crop_end = HUGE_VAL; + } else { + crop_end = atof(cropstr); + } + do_crop = true; + break; + } + + case 't': { + const char *cyclestr = strtok(optarg, ":"); + while (cyclestr != NULL) { + train_snap_points.push_back(atof(cyclestr)); + cyclestr = strtok(NULL, ":"); + } + do_train = true; + + // Set reasonable defaults (can be overridden later on the command line). + do_calibrate = false; + quiet = true; + break; + } + + case 'q': + quiet = true; + break; + case 'h': default: help(); @@ -185,30 +304,61 @@ void parse_options(int argc, char **argv) } } -int main(int argc, char **argv) +std::vector crop(const std::vector& pcm, float crop_start, float crop_end, int sample_rate) { - parse_options(argc, argv); + size_t start_sample, end_sample; + if (crop_start >= 0.0f) { + start_sample = std::min(lrintf(crop_start * sample_rate), pcm.size()); + } + if (crop_end >= 0.0f) { + end_sample = std::min(lrintf(crop_end * sample_rate), pcm.size()); + } + return std::vector(pcm.begin() + start_sample, pcm.begin() + end_sample); +} - make_lanczos_weight_table(); - std::vector pcm; - int sample_rate; - if (!read_audio_file(argv[optind], &pcm, &sample_rate)) { - exit(1); +// TODO: Support AVX here. +std::vector do_fir_filter(const std::vector& pcm, const float* filter) +{ + std::vector filtered_pcm; + filtered_pcm.reserve(pcm.size()); + for (unsigned i = NUM_FILTER_COEFF; i < pcm.size(); ++i) { + float s = 0.0f; + for (int j = 0; j < NUM_FILTER_COEFF; ++j) { + s += filter[j] * pcm[i - j]; + } + filtered_pcm.push_back(s); } -#if 0 - for (int i = 0; i < LEN; ++i) { - in[i] += rand() % 10000; + if (output_filtered) { + FILE *fp = fopen("filtered.raw", "wb"); + fwrite(filtered_pcm.data(), filtered_pcm.size() * sizeof(filtered_pcm[0]), 1, fp); + fclose(fp); } -#endif -#if 0 - for (int i = 0; i < LEN; ++i) { - printf("%d\n", in[i]); + return filtered_pcm; +} + +std::vector do_rc_filter(const std::vector& pcm, float freq, int sample_rate) +{ + std::vector filtered_pcm; + filtered_pcm.resize(pcm.size()); + Filter filter = Filter::hpf(2.0 * M_PI * freq / sample_rate); + for (unsigned i = 0; i < pcm.size(); ++i) { + filtered_pcm[i] = filter.update(pcm[i]); } -#endif - std::vector pulses; // in seconds + if (output_filtered) { + FILE *fp = fopen("filtered.raw", "wb"); + fwrite(filtered_pcm.data(), filtered_pcm.size() * sizeof(filtered_pcm[0]), 1, fp); + fclose(fp); + } + + return filtered_pcm; +} + +std::vector detect_pulses(const std::vector &pcm, int sample_rate) +{ + std::vector pulses; // Find the flanks. int last_bit = -1; @@ -239,7 +389,7 @@ int main(int argc, char **argv) } // down-flank! - double t = find_zerocrossing(pcm, i - 1) * (1.0 / sample_rate); + double t = find_zerocrossing(pcm, i - 1) * (1.0 / sample_rate) + crop_start; if (last_downflank > 0) { pulse p; p.time = t; @@ -250,6 +400,197 @@ int main(int argc, char **argv) } last_bit = bit; } + return pulses; +} + +void output_cycle_plot(const std::vector &pulses, double calibration_factor) +{ + FILE *fp = fopen("cycles.plot", "w"); + for (unsigned i = 0; i < pulses.size(); ++i) { + double cycles = pulses[i].len * calibration_factor * C64_FREQUENCY; + fprintf(fp, "%f %f\n", pulses[i].time, cycles); + } + fclose(fp); +} + +std::pair find_closest_point(double x, const std::vector &points) +{ + int best_point = 0; + double best_dist = (x - points[0]) * (x - points[0]); + for (unsigned j = 1; j < train_snap_points.size(); ++j) { + double dist = (x - points[j]) * (x - points[j]); + if (dist < best_dist) { + best_point = j; + best_dist = dist; + } + } + return std::make_pair(best_point, best_dist); +} + +float eval_badness(const std::vector& pulses, double calibration_factor) +{ + double sum_badness = 0.0; + for (unsigned i = 0; i < pulses.size(); ++i) { + double cycles = pulses[i].len * calibration_factor * C64_FREQUENCY; + if (cycles > 2000.0) cycles = 2000.0; // Don't make pauses arbitrarily bad. + std::pair selected_point_and_sq_dist = find_closest_point(cycles, train_snap_points); + sum_badness += selected_point_and_sq_dist.second; + } + return sqrt(sum_badness / (pulses.size() - 1)); +} + +void find_kmeans(const std::vector &pulses, double calibration_factor, const std::vector &initial_centers) +{ + std::vector last_centers = initial_centers; + std::vector sums; + std::vector num; + sums.resize(initial_centers.size()); + num.resize(initial_centers.size()); + for ( ;; ) { + for (unsigned i = 0; i < initial_centers.size(); ++i) { + sums[i] = 0.0f; + num[i] = 0; + } + for (unsigned i = 0; i < pulses.size(); ++i) { + double cycles = pulses[i].len * calibration_factor * C64_FREQUENCY; + // Ignore heavy outliers, which are almost always long pauses. + if (cycles > 2000.0) { + continue; + } + std::pair selected_point_and_sq_dist = find_closest_point(cycles, last_centers); + int p = selected_point_and_sq_dist.first; + sums[p] += cycles; + ++num[p]; + } + bool any_moved = false; + for (unsigned i = 0; i < initial_centers.size(); ++i) { + if (num[i] == 0) { + printf("K-means broke down, can't output new reference training points\n"); + return; + } + float new_center = sums[i] / num[i]; + if (fabs(new_center - last_centers[i]) > 1e-3) { + any_moved = true; + } + last_centers[i] = new_center; + } + if (!any_moved) { + break; + } + } + printf("New reference training points:"); + for (unsigned i = 0; i < last_centers.size(); ++i) { + printf(" %.3f", last_centers[i]); + } + printf("\n"); +} + +void spsa_train(const std::vector &pcm, int sample_rate) +{ + float filter[NUM_FILTER_COEFF] = { 1.0f }; // The rest is filled with 0. + + float start_c = INITIAL_C; + double best_badness = HUGE_VAL; + + for (int n = 1; n < NUM_ITER; ++n) { + float a = INITIAL_A * pow(n + A, -ALPHA); + float c = start_c * pow(n, -GAMMA); + + // find a random perturbation + float p[NUM_FILTER_COEFF]; + float filter1[NUM_FILTER_COEFF], filter2[NUM_FILTER_COEFF]; + for (int i = 0; i < NUM_FILTER_COEFF; ++i) { + p[i] = (rand() % 2) ? 1.0 : -1.0; + filter1[i] = std::max(std::min(filter[i] - c * p[i], 1.0f), -1.0f); + filter2[i] = std::max(std::min(filter[i] + c * p[i], 1.0f), -1.0f); + } + + std::vector pulses1 = detect_pulses(do_fir_filter(pcm, filter1), sample_rate); + std::vector pulses2 = detect_pulses(do_fir_filter(pcm, filter2), sample_rate); + float badness1 = eval_badness(pulses1, 1.0); + float badness2 = eval_badness(pulses2, 1.0); + + // Find the gradient estimator + float g[NUM_FILTER_COEFF]; + for (int i = 0; i < NUM_FILTER_COEFF; ++i) { + g[i] = (badness2 - badness1) / (2.0 * c * p[i]); + filter[i] -= a * g[i]; + filter[i] = std::max(std::min(filter[i], 1.0f), -1.0f); + } + if (badness2 < badness1) { + std::swap(badness1, badness2); + std::swap(filter1, filter2); + std::swap(pulses1, pulses2); + } + if (badness1 < best_badness) { + printf("\nNew best filter (badness=%f):", badness1); + for (int i = 0; i < NUM_FILTER_COEFF; ++i) { + printf(" %.5f", filter1[i]); + } + best_badness = badness1; + printf("\n"); + + find_kmeans(pulses1, 1.0, train_snap_points); + + if (output_cycles_plot) { + output_cycle_plot(pulses1, 1.0); + } + } + printf("%d ", n); + fflush(stdout); + } +} + +int main(int argc, char **argv) +{ + parse_options(argc, argv); + + make_lanczos_weight_table(); + std::vector pcm; + int sample_rate; + if (!read_audio_file(argv[optind], &pcm, &sample_rate)) { + exit(1); + } + + if (do_crop) { + pcm = crop(pcm, crop_start, crop_end, sample_rate); + } + + if (use_fir_filter) { + pcm = do_fir_filter(pcm, filter_coeff); + } + + if (use_rc_filter) { + pcm = do_rc_filter(pcm, rc_filter_freq, sample_rate); + } + + if (do_auto_level) { + pcm = level_samples(pcm, min_level, sample_rate); + if (output_leveled) { + FILE *fp = fopen("leveled.raw", "wb"); + fwrite(pcm.data(), pcm.size() * sizeof(pcm[0]), 1, fp); + fclose(fp); + } + } + +#if 0 + for (int i = 0; i < LEN; ++i) { + in[i] += rand() % 10000; + } +#endif + +#if 0 + for (int i = 0; i < LEN; ++i) { + printf("%d\n", in[i]); + } +#endif + + if (do_train) { + spsa_train(pcm, sample_rate); + exit(0); + } + + std::vector pulses = detect_pulses(pcm, sample_rate); double calibration_factor = 1.0; if (do_calibrate) { @@ -257,12 +598,7 @@ int main(int argc, char **argv) } if (output_cycles_plot) { - FILE *fp = fopen("cycles.plot", "w"); - for (unsigned i = 0; i < pulses.size(); ++i) { - double cycles = pulses[i].len * calibration_factor * C64_FREQUENCY; - fprintf(fp, "%f %f\n", pulses[i].time, cycles); - } - fclose(fp); + output_cycle_plot(pulses, calibration_factor); } output_tap(pulses, calibration_factor);