YangWX/crossdesk
1
0
Fork 0
mirror of https://github.com/kunkundi/crossdesk.git synced 2025-10-26 20:25:34 +08:00
Code Issues Packages Projects Releases Wiki Activity

Enable audio transmission

Browse Source
This commit is contained in:
dijunkun
2023-11-29 22:04:53 -08:00
parent 733434f9b3
commit 4a65a59803
12 changed files with 1520 additions and 271 deletions
Download Patch File Download Diff File Expand all files Collapse all files

83
test/audio_capture/audio_capture.cpp
View File

@@ -1,83 +0,0 @@
#include <stdio.h>
#ifdef _WIN32
// Windows
extern "C" {
#include <libavcodec/avcodec.h>
#include <libavdevice/avdevice.h>
#include <libavformat/avformat.h>
#include <libavutil/imgutils.h>
#include <libswscale/swscale.h>
};
#else
// Linux...
#ifdef __cplusplus
extern "C" {
#endif
#include <libavcodec/avcodec.h>
#include <libavdevice/avdevice.h>
#include <libavformat/avformat.h>
#include <libavutil/imgutils.h>
#include <libswscale/swscale.h>
#ifdef __cplusplus
};
#endif
#endif
int main(int argc, char **argv) {
int ret = 0;
char errors[1024] = {0};
// context
AVFormatContext *fmt_ctx = NULL; // ffmpeg<65>µġ<C2B5><C4A1>ļ<EFBFBD><C4BC><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD>
// paket
int count = 0;
AVPacket pkt;
// create file
char *out = "audio_old.pcm";
FILE *outfile = fopen(out, "wb+");
char *devicename = "default";
// register audio device
avdevice_register_all();
// get format
AVInputFormat *iformat = (AVInputFormat *)av_find_input_format("sndio");
// open audio
if ((ret = avformat_open_input(&fmt_ctx, devicename, iformat, NULL)) < 0) {
av_strerror(ret, errors, 1024);
printf("Failed to open audio device, [%d]%s\n", ret, errors);
return -1;
}
// <20><><EFBFBD><EFBFBD><EFBFBD><EFBFBD>Ƶ<EFBFBD><C6B5><EFBFBD><EFBFBD>Ϣ
if (avformat_find_stream_info(fmt_ctx, NULL) < 0) {
printf("111\n");
return -1;
}
// Ѱ<>ҵ<EFBFBD>һ<EFBFBD><D2BB><EFBFBD><EFBFBD>Ƶ<EFBFBD><C6B5><EFBFBD><EFBFBD><EFBFBD><EFBFBD>
int audioStreamIndex =
av_find_best_stream(fmt_ctx, AVMEDIA_TYPE_AUDIO, -1, -1, NULL, 0);
if (audioStreamIndex < 0) {
printf("222\n");
return -1;
}
av_init_packet(&pkt);
// read data form audio
while (ret = (av_read_frame(fmt_ctx, &pkt)) == 0 && count++ < 10000) {
av_log(NULL, AV_LOG_INFO, "pkt size is %d(%p), count=%d\n", pkt.size,
pkt.data, count);
fwrite(pkt.data, 1, pkt.size, outfile);
fflush(outfile);
av_packet_unref(&pkt); // release pkt
}
fclose(outfile);
avformat_close_input(&fmt_ctx); // releas ctx
return 0;
}

232
test/audio_capture/ffmpeg_audio.cpp Normal file
View File

@@ -0,0 +1,232 @@
extern "C" {
#include <libavcodec/avcodec.h>
#include <libavdevice/avdevice.h>
#include <libavfilter/avfilter.h>
#include <libavformat/avformat.h>
#include <libavutil/channel_layout.h>
#include <libavutil/imgutils.h>
#include <libavutil/opt.h>
#include <libavutil/samplefmt.h>
#include <libswresample/swresample.h>
#include <libswscale/swscale.h>
};
static int get_format_from_sample_fmt(const char **fmt,
enum AVSampleFormat sample_fmt) {
int i;
struct sample_fmt_entry {
enum AVSampleFormat sample_fmt;
const char *fmt_be, *fmt_le;
} sample_fmt_entries[] = {
{AV_SAMPLE_FMT_U8, "u8", "u8"},
{AV_SAMPLE_FMT_S16, "s16be", "s16le"},
{AV_SAMPLE_FMT_S32, "s32be", "s32le"},
{AV_SAMPLE_FMT_FLT, "f32be", "f32le"},
{AV_SAMPLE_FMT_DBL, "f64be", "f64le"},
};
*fmt = NULL;
for (i = 0; i < FF_ARRAY_ELEMS(sample_fmt_entries); i++) {
struct sample_fmt_entry *entry = &sample_fmt_entries[i];
if (sample_fmt == entry->sample_fmt) {
*fmt = AV_NE(entry->fmt_be, entry->fmt_le);
return 0;
}
}
fprintf(stderr, "Sample format %s not supported as output format\n",
av_get_sample_fmt_name(sample_fmt));
return AVERROR(EINVAL);
}
/**
* Fill dst buffer with nb_samples, generated starting from t. <20><><EFBFBD><EFBFBD>ģʽ<C4A3><CABD>
*/
static void fill_samples(double *dst, int nb_samples, int nb_channels,
int sample_rate, double *t) {
int i, j;
double tincr = 1.0 / sample_rate, *dstp = dst;
const double c = 2 * M_PI * 440.0;
/* generate sin tone with 440Hz frequency and duplicated channels */
for (i = 0; i < nb_samples; i++) {
*dstp = sin(c * *t);
for (j = 1; j < nb_channels; j++) dstp[j] = dstp[0];
dstp += nb_channels;
*t += tincr;
}
}
int main(int argc, char **argv) {
// <20><><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD>
int64_t src_ch_layout = AV_CH_LAYOUT_MONO;
int src_rate = 44100;
enum AVSampleFormat src_sample_fmt = AV_SAMPLE_FMT_DBL;
int src_nb_channels = 0;
uint8_t **src_data = NULL; // <20><><EFBFBD><EFBFBD>ָ<EFBFBD><D6B8>
int src_linesize;
int src_nb_samples = 1024;
// <20><><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD>
int64_t dst_ch_layout = AV_CH_LAYOUT_STEREO;
int dst_rate = 48000;
enum AVSampleFormat dst_sample_fmt = AV_SAMPLE_FMT_S16;
int dst_nb_channels = 0;
uint8_t **dst_data = NULL; // <20><><EFBFBD><EFBFBD>ָ<EFBFBD><D6B8>
int dst_linesize;
int dst_nb_samples;
int max_dst_nb_samples;
// <20><><EFBFBD><EFBFBD><EFBFBD>ļ<EFBFBD>
const char *dst_filename = NULL; // <20><><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD>pcm<63><6D><EFBFBD><EFBFBD><EFBFBD>أ<EFBFBD>Ȼ<EFBFBD>󲥷<EFBFBD><F3B2A5B7><EFBFBD>֤
FILE *dst_file;
int dst_bufsize;
const char *fmt;
// <20>ز<EFBFBD><D8B2><EFBFBD>ʵ<EFBFBD><CAB5>
struct SwrContext *swr_ctx;
double t;
int ret;
dst_filename = "res.pcm";
dst_file = fopen(dst_filename, "wb");
if (!dst_file) {
fprintf(stderr, "Could not open destination file %s\n", dst_filename);
exit(1);
}
// <20><><EFBFBD><EFBFBD><EFBFBD>ز<EFBFBD><D8B2><EFBFBD><EFBFBD><EFBFBD>
/* create resampler context */
swr_ctx = swr_alloc();
if (!swr_ctx) {
fprintf(stderr, "Could not allocate resampler context\n");
ret = AVERROR(ENOMEM);
goto end;
}
// <20><><EFBFBD><EFBFBD><EFBFBD>ز<EFBFBD><D8B2><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD>
/* set options */
// <20><><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD>
av_opt_set_int(swr_ctx, "in_channel_layout", src_ch_layout, 0);
av_opt_set_int(swr_ctx, "in_sample_rate", src_rate, 0);
av_opt_set_sample_fmt(swr_ctx, "in_sample_fmt", src_sample_fmt, 0);
// <20><><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD>
av_opt_set_int(swr_ctx, "out_channel_layout", dst_ch_layout, 0);
av_opt_set_int(swr_ctx, "out_sample_rate", dst_rate, 0);
av_opt_set_sample_fmt(swr_ctx, "out_sample_fmt", dst_sample_fmt, 0);
// <20><>ʼ<EFBFBD><CABC><EFBFBD>ز<EFBFBD><D8B2><EFBFBD>
/* initialize the resampling context */
if ((ret = swr_init(swr_ctx)) < 0) {
fprintf(stderr, "Failed to initialize the resampling context\n");
goto end;
}
/* allocate source and destination samples buffers */
// <20><><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD>Դ<EFBFBD><D4B4>ͨ<EFBFBD><CDA8><EFBFBD><EFBFBD><EFBFBD><EFBFBD>
src_nb_channels = av_get_channel_layout_nb_channels(src_ch_layout);
// <20><><EFBFBD><EFBFBD><EFBFBD><EFBFBD>Դ<EFBFBD><D4B4><EFBFBD><EFBFBD><EFBFBD>ڴ<EFBFBD><DAB4>ռ<EFBFBD>
ret = av_samples_alloc_array_and_samples(&src_data, &src_linesize,
src_nb_channels, src_nb_samples,
src_sample_fmt, 0);
if (ret < 0) {
fprintf(stderr, "Could not allocate source samples\n");
goto end;
}
/* compute the number of converted samples: buffering is avoided
* ensuring that the output buffer will contain at least all the
* converted input samples */
// <20><><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD>
max_dst_nb_samples = dst_nb_samples =
av_rescale_rnd(src_nb_samples, dst_rate, src_rate, AV_ROUND_UP);
/* buffer is going to be directly written to a rawaudio file, no alignment
*/
dst_nb_channels = av_get_channel_layout_nb_channels(dst_ch_layout);
// <20><><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD>ڴ<EFBFBD>
ret = av_samples_alloc_array_and_samples(&dst_data, &dst_linesize,
dst_nb_channels, dst_nb_samples,
dst_sample_fmt, 0);
if (ret < 0) {
fprintf(stderr, "Could not allocate destination samples\n");
goto end;
}
t = 0;
do {
/* generate synthetic audio */
// <20><><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD>Դ
fill_samples((double *)src_data[0], src_nb_samples, src_nb_channels,
src_rate, &t);
/* compute destination number of samples */
int64_t delay = swr_get_delay(swr_ctx, src_rate);
dst_nb_samples =
av_rescale_rnd(delay + src_nb_samples, dst_rate, src_rate, AV_ROUND_UP);
if (dst_nb_samples > max_dst_nb_samples) {
av_freep(&dst_data[0]);
ret = av_samples_alloc(dst_data, &dst_linesize, dst_nb_channels,
dst_nb_samples, dst_sample_fmt, 1);
if (ret < 0) break;
max_dst_nb_samples = dst_nb_samples;
}
// int fifo_size = swr_get_out_samples(swr_ctx,src_nb_samples);
// printf("fifo_size:%d\n", fifo_size);
// if(fifo_size < 1024)
// continue;
/* convert to destination format */
// ret = swr_convert(swr_ctx, dst_data, dst_nb_samples, (const
// uint8_t **)src_data, src_nb_samples);
ret = swr_convert(swr_ctx, dst_data, dst_nb_samples,
(const uint8_t **)src_data, src_nb_samples);
if (ret < 0) {
fprintf(stderr, "Error while converting\n");
goto end;
}
dst_bufsize = av_samples_get_buffer_size(&dst_linesize, dst_nb_channels,
ret, dst_sample_fmt, 1);
if (dst_bufsize < 0) {
fprintf(stderr, "Could not get sample buffer size\n");
goto end;
}
printf("t:%f in:%d out:%d\n", t, src_nb_samples, ret);
fwrite(dst_data[0], 1, dst_bufsize, dst_file);
} while (t < 10);
ret = swr_convert(swr_ctx, dst_data, dst_nb_samples, NULL, 0);
if (ret < 0) {
fprintf(stderr, "Error while converting\n");
goto end;
}
dst_bufsize = av_samples_get_buffer_size(&dst_linesize, dst_nb_channels, ret,
dst_sample_fmt, 1);
if (dst_bufsize < 0) {
fprintf(stderr, "Could not get sample buffer size\n");
goto end;
}
printf("flush in:%d out:%d\n", 0, ret);
fwrite(dst_data[0], 1, dst_bufsize, dst_file);
if ((ret = get_format_from_sample_fmt(&fmt, dst_sample_fmt)) < 0) goto end;
fprintf(stderr,
"Resampling succeeded. Play the output file with the command:\n"
"ffplay -f %s -channel_layout %" PRId64 " -channels %d -ar %d %s\n",
fmt, dst_ch_layout, dst_nb_channels, dst_rate, dst_filename);
end:
fclose(dst_file);
if (src_data) av_freep(&src_data[0]);
av_freep(&src_data);
if (dst_data) av_freep(&dst_data[0]);
av_freep(&dst_data);
swr_free(&swr_ctx);
return ret < 0;
}

53
test/audio_capture/output_audio.cpp Normal file
View File

@@ -0,0 +1,53 @@
#include <SDL2/SDL.h>
int main(int argc, char *argv[]) {
int ret;
SDL_AudioSpec wanted_spec, obtained_spec;
// Initialize SDL
if (SDL_Init(SDL_INIT_AUDIO) < 0) {
SDL_LogError(SDL_LOG_CATEGORY_APPLICATION, "Failed to initialize SDL: %s",
SDL_GetError());
return -1;
}
// Set audio format
wanted_spec.freq = 44100; // Sample rate
wanted_spec.format =
AUDIO_F32SYS; // Sample format (32-bit float, system byte order)
wanted_spec.channels = 2; // Number of channels (stereo)
wanted_spec.samples = 1024; // Buffer size (in samples)
wanted_spec.callback = NULL; // Audio callback function (not used here)
// Open audio device
ret = SDL_OpenAudio(&wanted_spec, &obtained_spec);
if (ret < 0) {
SDL_LogError(SDL_LOG_CATEGORY_APPLICATION,
"Failed to open audio device: %s", SDL_GetError());
return -1;
}
// Start playing audio
SDL_PauseAudio(0);
// Write PCM data to audio buffer
float *pcm_data = ...; // PCM data buffer (float, interleaved)
int pcm_data_size = ...; // Size of PCM data buffer (in bytes)
int bytes_written = SDL_QueueAudio(0, pcm_data, pcm_data_size);
// Wait until audio buffer is empty
while (SDL_GetQueuedAudioSize(0) > 0) {
SDL_Delay(100);
}
// Stop playing audio
SDL_PauseAudio(1);
// Close audio device
SDL_CloseAudio();
// Quit SDL
SDL_Quit();
return 0;
}

89
test/audio_capture/play_audio.cpp Normal file
View File

@@ -0,0 +1,89 @@
#include <SDL2/SDL.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
int main(int argc, char* argv[]) {
if (SDL_Init(SDL_INIT_AUDIO)) {
printf("SDL init error\n");
return -1;
}
// SDL_AudioSpec
SDL_AudioSpec wanted_spec;
SDL_zero(wanted_spec);
wanted_spec.freq = 48000;
wanted_spec.format = AUDIO_S16LSB;
wanted_spec.channels = 2;
wanted_spec.silence = 0;
wanted_spec.samples = 960;
wanted_spec.callback = NULL;
SDL_AudioDeviceID deviceID = 0;
// <20><><EFBFBD><EFBFBD><EFBFBD>豸
if ((deviceID = SDL_OpenAudioDevice(NULL, 0, &wanted_spec, NULL,
SDL_AUDIO_ALLOW_FREQUENCY_CHANGE)) < 2) {
printf("could not open audio device: %s\n", SDL_GetError());
// <20><><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD>е<EFBFBD><D0B5><EFBFBD>ϵͳ
SDL_Quit();
return 0;
}
SDL_PauseAudioDevice(deviceID, 0);
FILE* fp = nullptr;
fopen_s(&fp, "ls.pcm", "rb+");
if (fp == NULL) {
printf("cannot open this file\n");
return -1;
}
if (fp == NULL) {
printf("error \n");
}
Uint32 buffer_size = 4096;
char* buffer = (char*)malloc(buffer_size);
while (true) {
if (fread(buffer, 1, buffer_size, fp) != buffer_size) {
printf("end of file\n");
break;
}
SDL_QueueAudio(deviceID, buffer, buffer_size);
}
printf("Play...\n");
SDL_Delay(10000);
// Uint32 residueAudioLen = 0;
// while (true) {
// residueAudioLen = SDL_GetQueuedAudioSize(deviceID);
// printf("%10d\n", residueAudioLen);
// if (residueAudioLen <= 0) break;
// SDL_Delay(1);
// }
// while (true) {
// printf("1 <20><>ͣ 2 <20><><EFBFBD><EFBFBD> 3 <20>˳<EFBFBD> \n");
// int flag = 0;
// scanf_s("%d", &flag);
// if (flag == 1)
// SDL_PauseAudioDevice(deviceID, 1);
// else if (flag == 2)
// SDL_PauseAudioDevice(deviceID, 0);
// else if (flag == 3)
// break;
// }
SDL_CloseAudio();
SDL_Quit();
fclose(fp);
return 0;
}

225
test/audio_capture/play_loopback.cpp Normal file
View File

@@ -0,0 +1,225 @@
#include <SDL2/SDL.h>
#include <stdio.h>
#include <stdlib.h>
extern "C" {
#include <libavcodec/avcodec.h>
#include <libavdevice/avdevice.h>
#include <libavfilter/avfilter.h>
#include <libavformat/avformat.h>
#include <libavutil/channel_layout.h>
#include <libavutil/imgutils.h>
#include <libavutil/opt.h>
#include <libavutil/samplefmt.h>
#include <libswresample/swresample.h>
#include <libswscale/swscale.h>
};
static SDL_AudioDeviceID input_dev;
static SDL_AudioDeviceID output_dev;
static Uint8 *buffer = 0;
static int in_pos = 0;
static int out_pos = 0;
int64_t src_ch_layout = AV_CH_LAYOUT_MONO;
int src_rate = 48000;
enum AVSampleFormat src_sample_fmt = AV_SAMPLE_FMT_S16;
int src_nb_channels = 0;
uint8_t **src_data = NULL; // <20><><EFBFBD><EFBFBD>ָ<EFBFBD><D6B8>
int src_linesize;
int src_nb_samples = 480;
// <20><><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD>
int64_t dst_ch_layout = AV_CH_LAYOUT_MONO;
int dst_rate = 48000;
enum AVSampleFormat dst_sample_fmt = AV_SAMPLE_FMT_S16;
int dst_nb_channels = 0;
uint8_t **dst_data = NULL; // <20><><EFBFBD><EFBFBD>ָ<EFBFBD><D6B8>
int dst_linesize;
int dst_nb_samples;
int max_dst_nb_samples;
static unsigned char audio_buffer[960 * 3];
static int audio_len = 0;
// <20><><EFBFBD><EFBFBD><EFBFBD>ļ<EFBFBD>
const char *dst_filename = NULL; // <20><><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD>pcm<63><6D><EFBFBD><EFBFBD><EFBFBD>أ<EFBFBD>Ȼ<EFBFBD>󲥷<EFBFBD><F3B2A5B7><EFBFBD>֤
FILE *dst_file;
int dst_bufsize;
const char *fmt;
// <20>ز<EFBFBD><D8B2><EFBFBD>ʵ<EFBFBD><CAB5>
struct SwrContext *swr_ctx;
double t;
int ret;
char *out = "audio_old.pcm";
FILE *outfile = fopen(out, "wb+");
void cb_in(void *userdata, Uint8 *stream, int len) {
// If len < 4, the printf below will probably segfault
// SDL_QueueAudio(output_dev, stream, len);
int64_t delay = swr_get_delay(swr_ctx, src_rate);
dst_nb_samples =
av_rescale_rnd(delay + src_nb_samples, dst_rate, src_rate, AV_ROUND_UP);
if (dst_nb_samples > max_dst_nb_samples) {
av_freep(&dst_data[0]);
ret = av_samples_alloc(dst_data, &dst_linesize, dst_nb_channels,
dst_nb_samples, dst_sample_fmt, 1);
if (ret < 0) return;
max_dst_nb_samples = dst_nb_samples;
}
ret = swr_convert(swr_ctx, dst_data, dst_nb_samples,
(const uint8_t **)&stream, src_nb_samples);
if (ret < 0) {
fprintf(stderr, "Error while converting\n");
return;
}
dst_bufsize = av_samples_get_buffer_size(&dst_linesize, dst_nb_channels, ret,
dst_sample_fmt, 1);
if (dst_bufsize < 0) {
fprintf(stderr, "Could not get sample buffer size\n");
return;
}
printf("t:%f in:%d out:%d %d\n", t, src_nb_samples, ret, len);
memcpy(audio_buffer, dst_data[0], len);
// SDL_QueueAudio(output_dev, dst_data[0], len);
audio_len = len;
}
void cb_out(void *userdata, Uint8 *stream, int len) {
// If len < 4, the printf below will probably segfault
printf("cb_out len = %d\n", len);
SDL_memset(stream, 0, len);
if (audio_len == 0) return;
len = (len > audio_len ? audio_len : len);
SDL_MixAudioFormat(stream, audio_buffer, AUDIO_S16LSB, len,
SDL_MIX_MAXVOLUME);
}
int init() {
dst_filename = "res.pcm";
dst_file = fopen(dst_filename, "wb");
if (!dst_file) {
fprintf(stderr, "Could not open destination file %s\n", dst_filename);
exit(1);
}
// <20><><EFBFBD><EFBFBD><EFBFBD>ز<EFBFBD><D8B2><EFBFBD><EFBFBD><EFBFBD>
/* create resampler context */
swr_ctx = swr_alloc();
if (!swr_ctx) {
fprintf(stderr, "Could not allocate resampler context\n");
ret = AVERROR(ENOMEM);
return -1;
}
// <20><><EFBFBD><EFBFBD><EFBFBD>ز<EFBFBD><D8B2><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD>
/* set options */
// <20><><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD>
av_opt_set_int(swr_ctx, "in_channel_layout", src_ch_layout, 0);
av_opt_set_int(swr_ctx, "in_sample_rate", src_rate, 0);
av_opt_set_sample_fmt(swr_ctx, "in_sample_fmt", src_sample_fmt, 0);
// <20><><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD>
av_opt_set_int(swr_ctx, "out_channel_layout", dst_ch_layout, 0);
av_opt_set_int(swr_ctx, "out_sample_rate", dst_rate, 0);
av_opt_set_sample_fmt(swr_ctx, "out_sample_fmt", dst_sample_fmt, 0);
// <20><>ʼ<EFBFBD><CABC><EFBFBD>ز<EFBFBD><D8B2><EFBFBD>
/* initialize the resampling context */
if ((ret = swr_init(swr_ctx)) < 0) {
fprintf(stderr, "Failed to initialize the resampling context\n");
return -1;
}
/* allocate source and destination samples buffers */
// <20><><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD>Դ<EFBFBD><D4B4>ͨ<EFBFBD><CDA8><EFBFBD><EFBFBD><EFBFBD><EFBFBD>
src_nb_channels = av_get_channel_layout_nb_channels(src_ch_layout);
// <20><><EFBFBD><EFBFBD><EFBFBD><EFBFBD>Դ<EFBFBD><D4B4><EFBFBD><EFBFBD><EFBFBD>ڴ<EFBFBD><DAB4>ռ<EFBFBD>
ret = av_samples_alloc_array_and_samples(&src_data, &src_linesize,
src_nb_channels, src_nb_samples,
src_sample_fmt, 0);
if (ret < 0) {
fprintf(stderr, "Could not allocate source samples\n");
return -1;
}
/* compute the number of converted samples: buffering is avoided
* ensuring that the output buffer will contain at least all the
* converted input samples */
// <20><><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD>
max_dst_nb_samples = dst_nb_samples =
av_rescale_rnd(src_nb_samples, dst_rate, src_rate, AV_ROUND_UP);
/* buffer is going to be directly written to a rawaudio file, no alignment */
dst_nb_channels = av_get_channel_layout_nb_channels(dst_ch_layout);
// <20><><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD>ڴ<EFBFBD>
ret = av_samples_alloc_array_and_samples(&dst_data, &dst_linesize,
dst_nb_channels, dst_nb_samples,
dst_sample_fmt, 0);
if (ret < 0) {
fprintf(stderr, "Could not allocate destination samples\n");
return -1;
}
}
int main() {
init();
SDL_Init(SDL_INIT_AUDIO);
// 16Mb should be enough; the test lasts 5 seconds
buffer = (Uint8 *)malloc(16777215);
SDL_AudioSpec want_in, want_out, have_in, have_out;
SDL_zero(want_in);
want_in.freq = 48000;
want_in.format = AUDIO_S16LSB;
want_in.channels = 1;
want_in.samples = 480;
want_in.callback = cb_in;
input_dev = SDL_OpenAudioDevice(NULL, 1, &want_in, &have_in, 0);
printf("%d %d %d %d\n", have_in.freq, have_in.format, have_in.channels,
have_in.samples);
if (input_dev == 0) {
SDL_Log("Failed to open input: %s", SDL_GetError());
return 1;
}
SDL_zero(want_out);
want_out.freq = 48000;
want_out.format = AUDIO_S16LSB;
want_out.channels = 1;
want_out.samples = 480;
want_out.callback = cb_out;
output_dev = SDL_OpenAudioDevice(NULL, 0, &want_out, &have_out, 0);
printf("%d %d %d %d\n", have_out.freq, have_out.format, have_out.channels,
have_out.samples);
if (output_dev == 0) {
SDL_Log("Failed to open input: %s", SDL_GetError());
return 1;
}
SDL_PauseAudioDevice(input_dev, 0);
SDL_PauseAudioDevice(output_dev, 0);
while (1) {
}
SDL_CloseAudioDevice(output_dev);
SDL_CloseAudioDevice(input_dev);
free(buffer);
fclose(outfile);
}

95
test/audio_capture/resample.cpp Normal file
View File

@@ -0,0 +1,95 @@
#include <libavcodec/avcodec.h>
#include <libavformat/avformat.h>
#include <libswresample/swresample.h>
#include <stdint.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <unistd.h>
int main(int argc, char *argv[]) {
int ret;
AVFrame *frame = NULL;
AVFrame *resampled_frame = NULL;
AVCodecContext *codec_ctx = NULL;
SwrContext *swr_ctx = NULL;
// Initialize FFmpeg
av_log_set_level(AV_LOG_INFO);
av_register_all();
// Allocate input frame
frame = av_frame_alloc();
if (!frame) {
av_log(NULL, AV_LOG_ERROR, "Failed to allocate input frame\n");
return -1;
}
// Allocate output frame for resampled data
resampled_frame = av_frame_alloc();
if (!resampled_frame) {
av_log(NULL, AV_LOG_ERROR, "Failed to allocate output frame\n");
return -1;
}
// Set input frame properties
frame->format = AV_SAMPLE_FMT_FLTP; // Input sample format (float planar)
frame->channel_layout = AV_CH_LAYOUT_STEREO; // Input channel layout (stereo)
frame->sample_rate = 44100; // Input sample rate (44100 Hz)
frame->nb_samples = 1024; // Number of input samples
// Set output frame properties
resampled_frame->format =
AV_SAMPLE_FMT_S16; // Output sample format (signed 16-bit)
resampled_frame->channel_layout =
AV_CH_LAYOUT_STEREO; // Output channel layout (stereo)
resampled_frame->sample_rate = 48000; // Output sample rate (48000 Hz)
resampled_frame->nb_samples = av_rescale_rnd(
frame->nb_samples, resampled_frame->sample_rate, frame->sample_rate,
AV_ROUND_UP); // Number of output samples
// Initialize resampler context
swr_ctx = swr_alloc_set_opts(
NULL, av_get_default_channel_layout(resampled_frame->channel_layout),
av_get_default_sample_fmt(resampled_frame->format),
resampled_frame->sample_rate,
av_get_default_channel_layout(frame->channel_layout),
av_get_default_sample_fmt(frame->format), frame->sample_rate, 0, NULL);
if (!swr_ctx) {
av_log(NULL, AV_LOG_ERROR, "Failed to allocate resampler context\n");
return -1;
}
// Initialize and configure the resampler
if ((ret = swr_init(swr_ctx)) < 0) {
av_log(NULL, AV_LOG_ERROR, "Failed to initialize resampler context: %s\n",
av_err2str(ret));
return -1;
}
// Allocate buffer for output samples
ret = av_samples_alloc(resampled_frame->data, resampled_frame->linesize,
resampled_frame->channels, resampled_frame->nb_samples,
resampled_frame->format, 0);
if (ret < 0) {
av_log(NULL, AV_LOG_ERROR, "Failed to allocate output samples buffer: %s\n",
av_err2str(ret));
return -1;
}
// Resample the input data
ret = swr_convert(swr_ctx, resampled_frame->data, resampled_frame->nb_samples,
(const uint8_t **)frame->data, frame->nb_samples);
if (ret < 0) {
av_log(NULL, AV_LOG_ERROR, "Failed to resample input data: %s\n",
av_err2str(ret));
return -1;
}
// Cleanup and free resources
swr_free(&swr_ctx);
av_frame_free(&frame);
av_frame_free(&resampled_frame);
return 0;
}

196
test/audio_capture/sdl2_audio_capture.cpp
View File

@@ -2,6 +2,19 @@
#include <stdio.h>
#include <stdlib.h>
extern "C" {
#include <libavcodec/avcodec.h>
#include <libavdevice/avdevice.h>
#include <libavfilter/avfilter.h>
#include <libavformat/avformat.h>
#include <libavutil/channel_layout.h>
#include <libavutil/imgutils.h>
#include <libavutil/opt.h>
#include <libavutil/samplefmt.h>
#include <libswresample/swresample.h>
#include <libswscale/swscale.h>
};
static SDL_AudioDeviceID input_dev;
static SDL_AudioDeviceID output_dev;
@@ -9,54 +22,152 @@ static Uint8 *buffer = 0;
static int in_pos = 0;
static int out_pos = 0;
int64_t src_ch_layout = AV_CH_LAYOUT_MONO;
int src_rate = 48000;
enum AVSampleFormat src_sample_fmt = AV_SAMPLE_FMT_FLT;
int src_nb_channels = 0;
uint8_t **src_data = NULL; // <20><><EFBFBD><EFBFBD>ָ<EFBFBD><D6B8>
int src_linesize;
int src_nb_samples = 480;
// <20><><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD>
int64_t dst_ch_layout = AV_CH_LAYOUT_STEREO;
int dst_rate = 48000;
enum AVSampleFormat dst_sample_fmt = AV_SAMPLE_FMT_S16;
int dst_nb_channels = 0;
uint8_t **dst_data = NULL; // <20><><EFBFBD><EFBFBD>ָ<EFBFBD><D6B8>
int dst_linesize;
int dst_nb_samples;
int max_dst_nb_samples;
// <20><><EFBFBD><EFBFBD><EFBFBD>ļ<EFBFBD>
const char *dst_filename = NULL; // <20><><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD>pcm<63><6D><EFBFBD><EFBFBD><EFBFBD>أ<EFBFBD>Ȼ<EFBFBD>󲥷<EFBFBD><F3B2A5B7><EFBFBD>֤
FILE *dst_file;
int dst_bufsize;
const char *fmt;
// <20>ز<EFBFBD><D8B2><EFBFBD>ʵ<EFBFBD><CAB5>
struct SwrContext *swr_ctx;
double t;
int ret;
char *out = "audio_old.pcm";
FILE *outfile = fopen(out, "wb+");
void cb_in(void *userdata, Uint8 *stream, int len) {
// If len < 4, the printf below will probably segfault
{
fwrite(stream, 1, len, outfile);
fflush(outfile);
}
{
int64_t delay = swr_get_delay(swr_ctx, src_rate);
dst_nb_samples =
av_rescale_rnd(delay + src_nb_samples, dst_rate, src_rate, AV_ROUND_UP);
if (dst_nb_samples > max_dst_nb_samples) {
av_freep(&dst_data[0]);
ret = av_samples_alloc(dst_data, &dst_linesize, dst_nb_channels,
dst_nb_samples, dst_sample_fmt, 1);
if (ret < 0) return;
max_dst_nb_samples = dst_nb_samples;
}
// fwrite(stream, 1, len, outfile);
// fflush(outfile);
// SDL_memcpy(stream, buffer + in_pos, len);
// in_pos += len;
// printf("IN: %d\t%d %d %d %d\n", in_pos, stream[0], stream[1], stream[2],
// stream[3]);
ret = swr_convert(swr_ctx, dst_data, dst_nb_samples,
(const uint8_t **)&stream, src_nb_samples);
if (ret < 0) {
fprintf(stderr, "Error while converting\n");
return;
}
dst_bufsize = av_samples_get_buffer_size(&dst_linesize, dst_nb_channels,
ret, dst_sample_fmt, 1);
if (dst_bufsize < 0) {
fprintf(stderr, "Could not get sample buffer size\n");
return;
}
printf("t:%f in:%d out:%d\n", t, src_nb_samples, ret);
fwrite(dst_data[0], 1, dst_bufsize, dst_file);
}
}
void cb_out(void *userdata, Uint8 *stream, int len) {
// If len < 4, the printf below will probably segfault
fwrite(stream, 1, len, outfile);
fflush(outfile);
// if (out_pos >= in_pos) {
// // Output is way ahead of input; fill with emptiness
// memset(buffer + out_pos, 0, len * sizeof(Uint8));
// printf("OUT: %d\t(Empty)\n", out_pos);
// } else if (out_pos + len > in_pos) {
// // Output is reaching input; read until reaching input, and leave the
// rest
// // empty
// memset(buffer + out_pos, 0, len * sizeof(Uint8));
// SDL_memcpy(buffer + out_pos, stream, in_pos - out_pos);
// out_pos = in_pos;
// printf("OUT: %d\t%d %d %d %d (Partial)\n", out_pos, stream[0], stream[1],
// stream[2], stream[3]);
// } else {
// // Input is way ahead of output; read as much as requested
// SDL_memcpy(buffer + out_pos, stream, len);
// out_pos += len;
// printf("OUT: %d\t%d %d %d %d\n", out_pos, stream[0], stream[1],
// stream[2],
// stream[3]);
// }
SDL_memcpy(buffer + out_pos, stream, len);
out_pos += len;
}
// This is to make sure the output device works
// for (int i = 0; i < len; i++)
// stream[i] = (Uint8) random();
int init() {
dst_filename = "res.pcm";
dst_file = fopen(dst_filename, "wb");
if (!dst_file) {
fprintf(stderr, "Could not open destination file %s\n", dst_filename);
exit(1);
}
// <20><><EFBFBD><EFBFBD><EFBFBD>ز<EFBFBD><D8B2><EFBFBD><EFBFBD><EFBFBD>
/* create resampler context */
swr_ctx = swr_alloc();
if (!swr_ctx) {
fprintf(stderr, "Could not allocate resampler context\n");
ret = AVERROR(ENOMEM);
return -1;
}
// <20><><EFBFBD><EFBFBD><EFBFBD>ز<EFBFBD><D8B2><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD>
/* set options */
// <20><><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD>
av_opt_set_int(swr_ctx, "in_channel_layout", src_ch_layout, 0);
av_opt_set_int(swr_ctx, "in_sample_rate", src_rate, 0);
av_opt_set_sample_fmt(swr_ctx, "in_sample_fmt", src_sample_fmt, 0);
// <20><><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD>
av_opt_set_int(swr_ctx, "out_channel_layout", dst_ch_layout, 0);
av_opt_set_int(swr_ctx, "out_sample_rate", dst_rate, 0);
av_opt_set_sample_fmt(swr_ctx, "out_sample_fmt", dst_sample_fmt, 0);
// <20><>ʼ<EFBFBD><CABC><EFBFBD>ز<EFBFBD><D8B2><EFBFBD>
/* initialize the resampling context */
if ((ret = swr_init(swr_ctx)) < 0) {
fprintf(stderr, "Failed to initialize the resampling context\n");
return -1;
}
/* allocate source and destination samples buffers */
// <20><><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD>Դ<EFBFBD><D4B4>ͨ<EFBFBD><CDA8><EFBFBD><EFBFBD><EFBFBD><EFBFBD>
src_nb_channels = av_get_channel_layout_nb_channels(src_ch_layout);
// <20><><EFBFBD><EFBFBD><EFBFBD><EFBFBD>Դ<EFBFBD><D4B4><EFBFBD><EFBFBD><EFBFBD>ڴ<EFBFBD><DAB4>ռ<EFBFBD>
ret = av_samples_alloc_array_and_samples(&src_data, &src_linesize,
src_nb_channels, src_nb_samples,
src_sample_fmt, 0);
if (ret < 0) {
fprintf(stderr, "Could not allocate source samples\n");
return -1;
}
/* compute the number of converted samples: buffering is avoided
* ensuring that the output buffer will contain at least all the
* converted input samples */
// <20><><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD>
max_dst_nb_samples = dst_nb_samples =
av_rescale_rnd(src_nb_samples, dst_rate, src_rate, AV_ROUND_UP);
/* buffer is going to be directly written to a rawaudio file, no alignment */
dst_nb_channels = av_get_channel_layout_nb_channels(dst_ch_layout);
// <20><><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD>ڴ<EFBFBD>
ret = av_samples_alloc_array_and_samples(&dst_data, &dst_linesize,
dst_nb_channels, dst_nb_samples,
dst_sample_fmt, 0);
if (ret < 0) {
fprintf(stderr, "Could not allocate destination samples\n");
return -1;
}
}
int main() {
init();
SDL_Init(SDL_INIT_AUDIO);
// 16Mb should be enough; the test lasts 5 seconds
@@ -64,29 +175,18 @@ int main() {
SDL_AudioSpec want_in, want_out, have_in, have_out;
SDL_zero(want_out);
want_out.freq = 48000;
want_out.format = AUDIO_U16LSB;
want_out.channels = 2;
want_out.samples = 960;
want_out.callback = cb_out;
output_dev = SDL_OpenAudioDevice(NULL, 0, &want_out, &have_out,
SDL_AUDIO_ALLOW_ANY_CHANGE);
if (output_dev == 0) {
SDL_Log("Failed to open output: %s", SDL_GetError());
return 1;
}
SDL_zero(want_in);
want_in.freq = 48000;
want_in.format = AUDIO_U16LSB;
want_in.format = AUDIO_F32LSB;
want_in.channels = 2;
want_in.samples = 960;
want_in.callback = cb_in;
input_dev = SDL_OpenAudioDevice(NULL, 1, &want_in, &have_in,
SDL_AUDIO_ALLOW_ANY_CHANGE);
printf("%d %d %d %d\n", have_in.freq, have_in.format, have_in.channels,
have_in.samples);
if (input_dev == 0) {
SDL_Log("Failed to open input: %s", SDL_GetError());
return 1;

123
test/audio_capture/windows_capture.cpp Normal file
View File

@@ -0,0 +1,123 @@
#define __STDC_CONSTANT_MACROS
extern "C" {
#include <libavdevice/avdevice.h>
#include <libavformat/avformat.h>
#include <libavutil/log.h>
#include <libswresample/swresample.h>
}
#include <windows.h>
#include <memory>
#include <string>
#include <vector>
#pragma comment(lib, "avutil.lib")
#pragma comment(lib, "avdevice.lib")
#pragma comment(lib, "avformat.lib")
#pragma comment(lib, "avcodec.lib")
#pragma comment(lib, "Winmm.lib")
using std::shared_ptr;
using std::string;
using std::vector;
void capture_audio() {
// windows api <20><>ȡ<EFBFBD><C8A1>Ƶ<EFBFBD>豸<EFBFBD>б<EFBFBD><D0B1><EFBFBD>ffmpeg <20><><EFBFBD><EFBFBD>û<EFBFBD><C3BB><EFBFBD>ṩ<EFBFBD><E1B9A9>ȡ<EFBFBD><C8A1><EFBFBD><EFBFBD>Ƶ<EFBFBD>豸<EFBFBD><E8B1B8>api<70><69>
int nDeviceNum = waveInGetNumDevs();
vector<string> vecDeviceName;
for (int i = 0; i < nDeviceNum; ++i) {
WAVEINCAPS wic;
waveInGetDevCaps(i, &wic, sizeof(wic));
// ת<><D7AA>utf-8
int nSize = WideCharToMultiByte(CP_UTF8, 0, wic.szPname,
wcslen(wic.szPname), NULL, 0, NULL, NULL);
shared_ptr<char> spDeviceName(new char[nSize + 1]);
memset(spDeviceName.get(), 0, nSize + 1);
WideCharToMultiByte(CP_UTF8, 0, wic.szPname, wcslen(wic.szPname),
spDeviceName.get(), nSize, NULL, NULL);
vecDeviceName.push_back(spDeviceName.get());
av_log(NULL, AV_LOG_DEBUG, "audio input device : %s \n",
spDeviceName.get());
}
if (vecDeviceName.size() <= 0) {
av_log(NULL, AV_LOG_ERROR, "not find audio input device.\n");
return;
}
string sDeviceName = "audio=" + vecDeviceName[0]; // ʹ<>õ<EFBFBD>һ<EFBFBD><D2BB><EFBFBD><EFBFBD>Ƶ<EFBFBD>豸
// ffmpeg
avdevice_register_all(); // ע<><D7A2><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD>豸
AVInputFormat* ifmt =
(AVInputFormat*)av_find_input_format("dshow"); // <20><><EFBFBD>òɼ<C3B2><C9BC><EFBFBD>ʽ dshow
if (ifmt == NULL) {
av_log(NULL, AV_LOG_ERROR, "av_find_input_format for dshow fail.\n");
return;
}
AVFormatContext* fmt_ctx = NULL; // format <20><><EFBFBD><EFBFBD><EFBFBD><EFBFBD>
int ret = avformat_open_input(&fmt_ctx, sDeviceName.c_str(), ifmt,
NULL); // <20><><EFBFBD><EFBFBD><EFBFBD><EFBFBD>Ƶ<EFBFBD>豸
if (ret != 0) {
av_log(NULL, AV_LOG_ERROR, "avformat_open_input fail. return %d.\n", ret);
return;
}
AVPacket pkt;
int64_t src_rate = 44100;
int64_t dst_rate = 48000;
SwrContext* swr_ctx = swr_alloc();
uint8_t** dst_data = NULL;
int dst_linesize = 0;
// <20><><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD>
av_opt_set_int(swr_ctx, "in_channel_layout", AV_CH_LAYOUT_MONO, 0);
av_opt_set_int(swr_ctx, "in_sample_rate", src_rate, 0);
av_opt_set_sample_fmt(swr_ctx, "in_sample_fmt", AV_SAMPLE_FMT_S16, 0);
// <20><><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD>
av_opt_set_int(swr_ctx, "out_channel_layout", AV_CH_LAYOUT_STEREO, 0);
av_opt_set_int(swr_ctx, "out_sample_rate", dst_rate, 0);
av_opt_set_sample_fmt(swr_ctx, "out_sample_fmt", AV_SAMPLE_FMT_S16, 0);
// <20><>ʼ<EFBFBD><CABC>SwrContext
swr_init(swr_ctx);
FILE* fp = fopen("dst.pcm", "wb");
int count = 0;
while (count++ < 10) {
ret = av_read_frame(fmt_ctx, &pkt);
if (ret != 0) {
av_log(NULL, AV_LOG_ERROR, "av_read_frame fail, return %d .\n", ret);
break;
}
int out_samples_per_channel =
(int)av_rescale_rnd(1024, dst_rate, src_rate, AV_ROUND_UP);
int out_buffer_size = av_samples_get_buffer_size(
NULL, 1, out_samples_per_channel, AV_SAMPLE_FMT_S16, 0);
// uint8_t* out_buffer = (uint8_t*)av_malloc(out_buffer_size);
ret = av_samples_alloc_array_and_samples(
&dst_data, &dst_linesize, 2, out_buffer_size, AV_SAMPLE_FMT_S16, 0);
// <20><><EFBFBD><EFBFBD><EFBFBD>ز<EFBFBD><D8B2><EFBFBD>
swr_convert(swr_ctx, dst_data, out_samples_per_channel,
(const uint8_t**)&pkt.data, 1024);
fwrite(dst_data[1], 1, out_buffer_size, fp);
av_packet_unref(&pkt); // <20><><EFBFBD><EFBFBD><EFBFBD>ͷ<EFBFBD>pkt<6B><74><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD>ڴ棬<DAB4><E6A3AC><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD>ڴ<EFBFBD>й¶
}
fflush(fp); // ˢ<><CBA2><EFBFBD>ļ<EFBFBD>io<69><6F><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD><EFBFBD>
fclose(fp);
avformat_close_input(&fmt_ctx);
}
int main(int argc, char** argv) {
av_log_set_level(AV_LOG_DEBUG); // <20><><EFBFBD><EFBFBD>ffmpeg<65><67>־<EFBFBD><D6BE><EFBFBD>ȼ<EFBFBD>
capture_audio();
Sleep(1);
}