I trying to write ALSA application for recording audio, and when I try to set some parameters and then print them to the screen I getting some default numbers that i cant change
#include <alsa/asoundlib.h>
using namespace std;
typedef struct {
int audio;
int recording;
void *cons;
snd_pcm_t *inhandle;
snd_pcm_t *outhandle;
unsigned long sampleIndex;
unsigned long inlen;
unsigned long sampleRate;
} audio_t;
static audio_t aud;
void aboutAlsa(snd_pcm_t *handle,snd_pcm_hw_params_t *params) {
unsigned int val, val2;
snd_pcm_format_t val3;
int dir;
snd_pcm_uframes_t frames;
printf("ALSA library version: %s\n",SND_LIB_VERSION_STR);
printf("PCM handle name = '%s'\n",snd_pcm_name(handle));
printf("PCM state = %s\n",snd_pcm_state_name(snd_pcm_state(handle)));
snd_pcm_hw_params_get_access(params,(snd_pcm_access_t *) &val);
printf("access type = %s\n",snd_pcm_access_name((snd_pcm_access_t)val));
snd_pcm_hw_params_get_format(params, &val3);
printf("format = '%s' (%s)\n",snd_pcm_format_name(val3),
snd_pcm_format_description(val3));
snd_pcm_hw_params_get_subformat(params,(snd_pcm_subformat_t *)&val);
printf("subformat = '%s' (%s)\n",snd_pcm_subformat_name((snd_pcm_subformat_t)val),
snd_pcm_subformat_description((snd_pcm_subformat_t)val));
snd_pcm_hw_params_get_channels(params, &val);
printf("channels = %d\n", val);
snd_pcm_hw_params_get_rate(params, &val, &dir);
printf("rate = %d bps\n", val);
snd_pcm_hw_params_get_period_time(params,&val, &dir);
printf("period time = %d us\n", val);
snd_pcm_hw_params_get_period_size(params,&frames, &dir);
printf("period size = %d frames\n", (int)frames);
snd_pcm_hw_params_get_buffer_time(params,&val, &dir);
printf("buffer time = %d us\n", val);
snd_pcm_hw_params_get_buffer_size(params,(snd_pcm_uframes_t *) &val);
printf("buffer size = %d frames\n", val);
snd_pcm_hw_params_get_periods(params, &val, &dir);
printf("periods per buffer = %d frames\n", val);
snd_pcm_hw_params_get_rate_numden(params,&val, &val2);
printf("exact rate = %d/%d bps\n", val, val2);
val = snd_pcm_hw_params_get_sbits(params);
printf("significant bits = %d\n", val);
return;
}
static int openKnownAudio(int record) {
int rc;
int SAMPLERATE = 16000;
unsigned int val;
int dir=0;
snd_pcm_t *handle;
snd_pcm_hw_params_t *hw_params=NULL;
snd_pcm_uframes_t frames;
size_t esz = 256;
char err[esz];
/* Open PCM device for recording (capture). */
if (record) {
if ((rc=snd_pcm_open(&aud.inhandle, "default",SND_PCM_STREAM_CAPTURE, 0))<0) {
snprintf(err, esz, "unable to open pcm device for recording: %s\n",snd_strerror(rc));
}
handle=aud.inhandle;
} else {
if ((rc=snd_pcm_open(&aud.outhandle, "default",SND_PCM_STREAM_PLAYBACK, 0))<0) {
snprintf(err, esz, "unable to open pcm device for playback: %s\n",snd_strerror(rc));
}
handle=aud.outhandle;
}
/* Configure hardware parameters */
if((rc=snd_pcm_hw_params_malloc(&hw_params)) < 0) {
snprintf(err, esz, "unable to malloc hw_params: %s\n",snd_strerror(rc));
}
if((rc=snd_pcm_hw_params_any(handle, hw_params))<0) {
snprintf(err, esz, "unable to setup hw_params: %s\n",snd_strerror(rc));
}
if((rc=snd_pcm_hw_params_set_access(handle, hw_params, SND_PCM_ACCESS_RW_INTERLEAVED))<0) {
snprintf(err, esz, "unable to set access mode: %s\n",snd_strerror(rc));
}
if((rc=snd_pcm_hw_params_set_format(handle, hw_params, SND_PCM_FORMAT_S16_LE))<0) {
snprintf(err, esz, "unable to set format: %s\n",snd_strerror(rc));
}
if((rc=snd_pcm_hw_params_set_channels(handle, hw_params, 1))<0) {
snprintf(err, esz, "unable to set channels: %s\n",snd_strerror(rc));
}
val = SAMPLERATE;
dir = 0;
if((rc=snd_pcm_hw_params_set_rate(handle, hw_params, SAMPLERATE,0))<0) {
snprintf(err, esz, "unable to set samplerate: %s\n",snd_strerror(rc));
}
if (val!=SAMPLERATE) {
snprintf(err, esz, "unable to set requested samplerate: requested=%i got=%i\n",SAMPLERATE,val);
}
frames = 64;
if ((rc=snd_pcm_hw_params_set_period_size_near(handle,hw_params, &frames, &dir))<0) {
snprintf(err, esz, "unable to set period size: %s\n",snd_strerror(rc));
}
frames = 4096;
if ((rc=snd_pcm_hw_params_set_buffer_size_near(handle,hw_params, &frames))<0) {
snprintf(err, esz, "unable to set buffer size: %s\n",snd_strerror(rc));
}
if ((rc = snd_pcm_hw_params(handle, hw_params))<0) {
snprintf(err, esz, "unable to set hw parameters: %s\n",snd_strerror(rc));
}
aboutAlsa(handle,hw_params);
snd_pcm_hw_params_free(hw_params);
aud.recording = (record)? 1:0;
aud.audio=1;
return 1;
}
This what I get on raspberry pi when I run it:
ALSA library version: 1.0.28
PCM handle name = 'default'
PCM state = PREPARED
access type = RW_INTERLEAVED
format = 'S16_LE' (Signed 16 bit Little Endian)
subformat = 'STD' (Standard)
channels = 1
rate = 16000 bps
period time = 21333 us
period size = 341 frames
buffer time = 256000 us
buffer size = 4096 frames
periods per buffer = 4096 frames
exact rate = 16000/1 bps
significant bits = 16
And this is what I get when I run it on desktop pc:
ALSA library version: 1.0.28
PCM handle name = 'default'
PCM state = PREPARED
access type = RW_INTERLEAVED
format = 'S16_LE' (Signed 16 bit Little Endian)
subformat = 'STD' (Standard)
channels = 1
rate = 16000 bps
period time = 4000 us
period size = 64 frames
buffer time = 256000 us
buffer size = 4096 frames
periods per buffer = 64 frames
exact rate = 16000/1 bps
significant bits = 16
As you can see I'm trying to set the period size to 64 and getting back 341, this value only changes when I change the rate, lets say I set the rate to 44100 and this what I getting back:
rate = 44100 bps
period time = 21333 us
period size = 940 frames
buffer time = 85328 us
buffer size = 3763 frames
periods per buffer = 3763 frames
On desktop pc this doesn't happens I tried to trace down this functions in alsa-lib but I getting lost there also tried different sound cards and still getting same result .
In case of PulseAudio you did set the PulseAudio device , not the real device.
The real HW can have the limitation, you must correctly react to.
If you'd like to see min/max boundary of some parameter, you can do the next:
using snd_pcm_hw_params_dump function
snd_pcm_hw_params_t *params;
snd_pcm_t *pcm_handle;
int pcm;
/* Open the PCM device in playback mode */
pcm = snd_pcm_open(&pcm_handle, PCM_DEVICE, SND_PCM_STREAM_PLAYBACK, 0);
if (pcm < 0) {
printf("ERROR: Can't open \"%s\" PCM device. %s\n", PCM_DEVICE, snd_strerror(pcm));
goto error_handling;
}
/* Allocate parameters object and fill it with default values*/
snd_pcm_hw_params_alloca(¶ms);
pcm = snd_pcm_hw_params_any(pcm_handle, params);
if (pcm < 0) {
printf("Broken configuration for this PCM: no configurations available\n");
goto error_handling;
}
printf("hw boundary params ***********************\n");
snd_pcm_hw_params_dump(params, log);
printf("*******************************************\n");
The same using min/max functions
snd_pcm_t* pcm;
snd_pcm_hw_params_t* hw_parameters;
int parameter;
//... open device and allocate hw params here
/*Fill params with a full configuration space for a PCM.
The configuration space will be filled with all possible
ranges for the PCM device.*/
snd_pcm_hw_params_any(pcm,hw_parameters);
/* please substitute <parameter name> with real parameter name
for example buffer_size, buffer_time, rate, etc*/
snd_pcm_hw_params_get_<parameter name>_min(hw_parameters,¶meter);
printf("<parameter name> min : %d/n", parameter);
snd_pcm_hw_params_get_<parameter name>_max(hw_parameters,¶meter);
printf("<parameter name> max : %d/n", parameter);
I faced with the same issue, when tried to set the period size.
There are my boundary (two different pcm devices):
log #1
hw boundary params ***********************
ACCESS: RW_INTERLEAVED
FORMAT: U8 S16_LE S16_BE S24_LE S24_BE S32_LE S32_BE FLOAT_LE FLOAT_BE MU_LAW A_LAW S24_3LE S24_3BE
SUBFORMAT: STD
SAMPLE_BITS: [8 32]
FRAME_BITS: [8 1024]
CHANNELS: [1 32]
RATE: [1 192000]
PERIOD_TIME: (5 4294967295)
PERIOD_SIZE: [1 1398102)
PERIOD_BYTES: [128 1398102)
PERIODS: [3 1024]
BUFFER_TIME: (15 4294967295]
BUFFER_SIZE: [3 4194304]
BUFFER_BYTES: [384 4194304]
TICK_TIME: ALL
*******************************************
log#2
**********************************DEBUG
period time min : 21333
period time max : 21334
buffer time min : 1
buffer time max : -1
channels min : 1
channels max : 10000
rate min : 4000
rate max : -1
period size min : 85
period size max : 91628833
buffer size min : 170
buffer size max : 274877906
**********************************DEBUG_END
Here we can't change the period size due to period time limitation.
Related
I'm using a USB sound card to create a real-time Linux player. The APP is developed based on ALSA's API.
I found that the driver's buffer is too large. I don't know how to reduce the buffer size via userspace.
For example, when I initialize ALSA by 16-bit stereo 44100Hz, the total buffer size will be 256 frames(snd_pcm_hw_params_get_period_size) * 1024 periods(snd_pcm_hw_params_get_periods). So that the total buffer time is 256*1024/44100=5.94(seconds). That's too large for my REAL-TIME use case!
my exp 01: I've been trying to reduce to periods counts by snd_pcm_hw_params_test_periods, but the driver's buffer size (frames*periods) is ALL THE SAME!
here's my sample code
bool init(const std::string &PCM_DEVICE, const int channels, const int rate) {
unsigned int pcm;
unsigned int tmp;
snd_pcm_hw_params_t* params = nullptr;
/* Open the PCM device in playback mode */
if (pcm = snd_pcm_open(&pcm_handle_, PCM_DEVICE.c_str(), SND_PCM_STREAM_PLAYBACK, 0) < 0) {
printf("ERROR: Can't open \"%s\" PCM device. %s\n", PCM_DEVICE, snd_strerror(pcm));
return false;
}
/* Allocate parameters object and fill it with default values*/
snd_pcm_hw_params_alloca(¶ms);
snd_pcm_hw_params_any(pcm_handle_, params);
/* Set parameters */
if (pcm = snd_pcm_hw_params_set_access(pcm_handle_, params,
SND_PCM_ACCESS_RW_INTERLEAVED) < 0)
printf("ERROR: Can't set interleaved mode. %s\n", snd_strerror(pcm));
if (pcm = snd_pcm_hw_params_set_format(pcm_handle_, params,
SND_PCM_FORMAT_S16_LE) < 0)
printf("ERROR: Can't set format. %s\n", snd_strerror(pcm));
if (pcm = snd_pcm_hw_params_set_channels(pcm_handle_, params, channels) < 0)
printf("ERROR: Can't set channels number. %s\n", snd_strerror(pcm));
if (pcm = snd_pcm_hw_params_set_rate_near(pcm_handle_, params, (unsigned int*)&rate, 0) < 0)
printf("ERROR: Can't set rate. %s\n", snd_strerror(pcm));
/* Write parameters */
if (pcm = snd_pcm_hw_params(pcm_handle_, params) < 0)
printf("ERROR: Can't set harware parameters. %s\n", snd_strerror(pcm));
/* Resume information */
printf("PCM name: '%s'\n", snd_pcm_name(pcm_handle_));
printf("PCM state: %s\n", snd_pcm_state_name(snd_pcm_state(pcm_handle_)));
snd_pcm_hw_params_get_channels(params, &tmp);
printf("channels: %i ", tmp);
if (tmp == 1)
printf("(mono)\n");
else if (tmp == 2)
printf("(stereo)\n");
snd_pcm_hw_params_get_rate(params, &tmp, 0);
printf("rate: %d bps\n", tmp);
/* Allocate buffer to hold single period */
snd_pcm_hw_params_get_period_size(params, &frames_, 0);
LOGI << "frames_ = " << frames_;
buffSize_ = frames_ * channels * 2 /* 2 -> sample size */;
buff_.resize(buffSize_);
snd_pcm_hw_params_get_period_time(params, &tmp, NULL);
snd_pcm_hw_params_get_periods(params, &tmp, NULL); // https://www.alsa-project.org/wiki/FramesPeriods
LOGI << "periods_ =" << tmp;
return true;
}
I am trying to record audio from an INMP441 which is connected to a ESP32 but returning the buffer containing the bytes the microphone read always leads to something which is NULL.
The code for setting up i2s and the microphone is this:
// i2s config
const i2s_config_t i2s_config = {
.mode = i2s_mode_t(I2S_MODE_MASTER | I2S_MODE_RX), // receive
.sample_rate = SAMPLE_RATE, // 44100 (44,1KHz)
.bits_per_sample = I2S_BITS_PER_SAMPLE_32BIT, // 32 bits per sample
.channel_format = I2S_CHANNEL_FMT_ONLY_LEFT, // use right channel
.communication_format = i2s_comm_format_t(I2S_COMM_FORMAT_I2S | I2S_COMM_FORMAT_I2S_MSB),
.intr_alloc_flags = ESP_INTR_FLAG_LEVEL1, // interrupt level 1
.dma_buf_count = 64, // number of buffers
.dma_buf_len = SAMPLES_PER_BUFFER}; // 512
// pin config
const i2s_pin_config_t pin_config = {
.bck_io_num = gpio_sck, // serial clock, sck (gpio 33)
.ws_io_num = gpio_ws, // word select, ws (gpio 32)
.data_out_num = I2S_PIN_NO_CHANGE, // only used for speakers
.data_in_num = gpio_sd // serial data, sd (gpio 34)
};
// config i2s driver and pins
// fct must be called before any read/write
esp_err_t err = i2s_driver_install(I2S_PORT, &i2s_config, 0, NULL);
if (err != ESP_OK)
{
Serial.printf("Failed installing the driver: %d\n", err);
}
err = i2s_set_pin(I2S_PORT, &pin_config);
if (err != ESP_OK)
{
Serial.printf("Failed setting pin: %d\n", err);
}
Serial.println("I2S driver installed! :-)");
Setting up the i2s stuff is no problem at all. The tricky part for me is reading from the i2s:
// 44KHz * Byte per sample * time in seconds = total size in bytes
const size_t recordSize = (SAMPLE_RATE * I2S_BITS_PER_SAMPLE_32BIT / 8) * recordTime; //recordTime = 5s
// size in bytes
size_t totalReadSize = 0;
// 32 bits per sample set in config * 1024 samples per buffers = total bits per buffer
char *samples = (char *)calloc(totalBitsPerBuffer, sizeof(char));
// number of bytes read
size_t bytesRead;
Serial.println("Start recording...");
// read until wanted size is reached
while (totalReadSize < recordSize)
{
// read to buffer
esp_err_t err = i2s_read(I2S_PORT, (void *)samples, totalBitsPerBuffer, &bytesRead, portMAX_DELAY);
// check if error occurd, if so stop recording
if (err != ESP_OK)
{
Serial.println("Error while recording!");
break;
}
// check if bytes read works → yes
/*
for (int i = 0; i < bytesRead; i++)
{
uint8_t sample = (uint8_t) samples[i];
Serial.print(sample);
} */
// add read size to total read size
totalReadSize += bytesRead;
// Serial.printf("Currently recorded %d%% \n", totalReadSize * 100 / recordSize);
}
// convert bytes to mb
double_t totalReadSizeMB = (double_t)totalReadSize / 1e+6;
Serial.printf("Total read size: %fMb\n", totalReadSizeMB);
Serial.println("Samples deref");
Serial.println(*samples);
Serial.println("Samples");
Serial.println(samples);
return samples;
Using this code leads to the following output:
I2S driver installed! :-)
Start recording...
Total read size: 0.884736Mb
Samples deref
␀
Samples
When I uncomment the part where I iterate over the bytes read part I get something like this:
200224231255255224210022418725525522493000902552550238002241392542552241520020425225508050021624525501286700194120022461104022421711102242271030018010402242510000188970224141930022291022410185022487830021679001127500967200666902241776600246610224895902244757022418353002224802242274302249741022419339009435001223102242432602243322022412120001241402245911022418580084402248325525522461252255044249255224312452552242212372552241272352550342302552241212262552242112212550252216255014621325501682092550112205255224161202255224237198255224235194255224231922552248518725501141832550421812552241951762550144172255018168255034164255224173157255018215525522455152255028148255021014425505214025522487137255014613225522412112825502361252550180120255018011725522451172550252113255224133111255061082550248105255224891042552249910125522439972550138942552242279225503287255224101832552242478125522410178255224231732552244970255224336525501766225501426125502325625522424553255224109492550186[...]
This shows that the microphone is able to record, but I cant return the actual value of the buffer.
While programming this code I looked up at the official doku and some code which seems to work elsewhere.
I am also new to C++ and am not used to work with pointers.
Does anyone know what the problem could be?
I'm using FFmpeg and try to encode and decode a raw PCM sound to Opus using a built-in FFmpeg "opus" codec. My input samples are raw PCM 8000 Hz 16 bit mono, in AV_SAMPLE_FMT_S16 format. Since Opus requires sample format AV_SAMPLE_FMT_FLTP and sample rate 48000 Hz only, so I resample my samples before encode them.
I have two instances of ResamplerAudio class that does the work of resampling audio samples and has a member of SwrContext, I use the first instance of ResamplerAudio for resampling a raw PCM input audio before encoding and the second for resampling decoded audio to get it's format and sample rate the same as source values of input raw audio.
ResamplerAudio class has a function that init it's SwrContext member like this:
void ResamplerAudio::init(AVCodecContext *codecContext, int inSampleRate, int outSampleRate, AVSampleFormat inSampleFmt, AVSampleFormat outSampleFmt)
{
swrContext = swr_alloc();
if (!swrContext)
{
LOGE(TAG, "[init] Couldn't allocate swr context");
return;
}
av_opt_set_int(swrContext, "in_channel_layout", (int64_t) codecContext->channel_layout, 0);
av_opt_set_int(swrContext, "out_channel_layout", (int64_t) codecContext->channel_layout, 0);
av_opt_set_int(swrContext, "in_channel_count", codecContext->channels, 0);
av_opt_set_int(swrContext, "out_channel_count", codecContext->channels, 0);
av_opt_set_int(swrContext, "in_sample_rate", inSampleRate, 0);
av_opt_set_int(swrContext, "out_sample_rate", outSampleRate, 0);
av_opt_set_sample_fmt(swrContext, "in_sample_fmt", inSampleFmt, 0);
av_opt_set_sample_fmt(swrContext, "out_sample_fmt", outSampleFmt, 0);
int ret = swr_init(swrContext);
if (ret < 0)
{
LOGE(TAG, "[init] swr_init error: %s", av_err2str(ret));
return;
}
LOGD(TAG, "[init] success codecContext->channel_layout: %d; inSampleRate: %d; outSampleRate: %d; inSampleFmt: %d; outSampleFmt: %d", (int) codecContext->channel_layout, inSampleRate, outSampleRate, inSampleFmt, outSampleFmt);
}
And I call ResamplerAudio::init function for the first instance of ResamplerAudio (this instance do resamping a raw PCM input audio before encoding and I called it resamplerEncoder) with the following args:
resamplerEncoder->init(contextEncoder, 8000, 48000, AV_SAMPLE_FMT_S16, AV_SAMPLE_FMT_FLTP);
The second instance of ResamplerAudio (this instance do resamping after decoding audio from Opus and I called it resamplerDecoder) I init with the following args:
resamplerDecoder->init(contextDecoder, 48000, 8000, AV_SAMPLE_FMT_FLTP, AV_SAMPLE_FMT_S16);
The function of ResamplerAudio that does resampling looks like this:
std::vector<uint8_t> ResamplerAudio::convert(uint8_t **inData, int inSamplesCount, int outChannels, int outFormat)
{
std::vector<uint8_t> result;
uint8_t *dstData = NULL;
const int dstNbSamples = swr_get_out_samples(swrContext, inSamplesCount);
av_samples_alloc(&dstData, NULL, outChannels, dstNbSamples, AVSampleFormat(outFormat), 1);
int resampledSize = swr_convert(swrContext, &dstData, dstNbSamples, (const uint8_t **)inData, inSamplesCount);
int dstBufSize = av_samples_get_buffer_size(NULL, outChannels, resampledSize, AVSampleFormat(outFormat), 1);
if (dstBufSize <= 0) return result;
std::copy(&dstData[0], &dstData[dstBufSize], std::back_inserter(result));
return result;
}
And I call ResamplerAudio::convert function before encoding with the following args:
// data - an array of raw pcm audio
// dataLength - the length of data array
// getSamplesCount() - function that calculates samples count
// frameEncode - AVFrame that using for encode audio
std::vector<uint8_t> resampledData = resamplerEncoder->convert(&data, getSamplesCount(dataLength, frameEncode->channels, AV_SAMPLE_FMT_S16), frameEncode->channels, frameEncode->format);
getSamplesCount() function looks like this:
getSamplesCount(int bytesCount, int channels, AVSampleFormat format)
{
return bytesCount / av_get_bytes_per_sample(format) / channels;
}
After that I fill my frameEncode with resampled samples:
memcpy(&frame->data[0][0], &resampledData[0], sizeof(uint8_t) * resampledDataLength);
And pass frameEncode to encoding like this encodeFrame(resampledDataLength):
void encodeFrame(int dataLength)
{
/* send the frame for encoding */
int ret = avcodec_send_frame(contextEncoder, frameEncode);
if (ret < 0)
{
LOGE(TAG, "[encodeFrame] avcodec_send_frame error: %s", av_err2str(ret));
return;
}
/* read all the available output packets (in general there may be any number of them */
while (ret >= 0)
{
ret = avcodec_receive_packet(contextEncoder, packetEncode);
if (ret < 0 && ret != AVERROR(EAGAIN)) LOGE(TAG, "[encodeFrame] error in avcodec_receive_packet: %s", av_err2str(ret));
if (ret < 0) break;
// encodedData - std::vector<uint8_t> that stores encoded data
std::copy(&packetEncode->data[0], &packetEncode->data[dataLength], std::back_inserter(encodedData));
av_packet_unref(packetEncode);
}
}
Then I decode my encoded samples and do resampling to get back them in source sample format and sample rate so I call ResamplerAudio::convert function for resamplerDecoder with the following args:
// frameDecode - AVFrame that holds decoded audio
std::vector<uint8_t> resampledData = resamplerDecoder->convert(frameDecode->data, frameDecode->nb_samples, frameDecode->channels, AV_SAMPLE_FMT_S16);
And result sound is choppy and I also noticed that the decoded array size is bigger than the source array size with raw pcm audio.
Please any ideas what I'm doing wrong?
UPD 18.05.2020
I tested my resampling logic, I did resampling of raw pcm sound without any encoding and decoding routines. First I tried to convert the sample rate of input sound from 8000 Hz to 48000 Hz than I took resampled samples from step above and convert it's sample rate from 48000 Hz to 8000 Hz and the result sound is perfect and clean, also I did the same steps but I converted not a sample rate but a sample format from AV_SAMPLE_FMT_S16 to AV_SAMPLE_FMT_FLTP and vice versa and again the result sound is perfect and clean, also I got the same result when I coverted both a sample rate and a sample format.
So I assume that the problem of distorted and choppy sound is in my encoding or decoding routine, I think most likely in decoding routine because after decoding I ALWAYS get AVFrame with 960 nb_samples despite what was the size of input sound.
My decoding routine looks like this:
std::vector<uint8_t> decode(uint8_t *data, unsigned int dataLength)
{
decodedData.clear();
int dataSize = dataLength;
while (dataSize > 0)
{
if (!frameDecode)
{
frameDecode = av_frame_alloc();
if (!frameDecode)
{
LOGE(TAG, "[decode] Couldn't allocate the frame");
return EMPTY_DATA;
}
}
ret = av_parser_parse2(parser, contextDecoder, &packetDecode->data, &packetDecode->size, &data[0], dataSize, AV_NOPTS_VALUE, AV_NOPTS_VALUE, 0);
if (ret < 0) {
LOGE(TAG, "[decode] av_parser_parse2 error: %s", av_err2str(ret));
return EMPTY_DATA;
}
data += ret;
dataSize -= ret;
doDecode();
}
return decodedData;
}
void doDecode()
{
if (packetDecode->size) {
/* send the packet with the compressed data to the decoder */
int ret = avcodec_send_packet(contextDecoder, packetDecode);
if (ret < 0) LOGE(TAG, "[decode] avcodec_send_packet error: %s", av_err2str(ret));
/* read all the output frames (in general there may be any number of them */
while (ret >= 0)
{
ret = avcodec_receive_frame(contextDecoder, frameDecode);
if (ret < 0 && ret != AVERROR(EAGAIN) && ret != AVERROR_EOF) LOGE(TAG, "[decode] avcodec_receive_frame error: %s", av_err2str(ret));
if (ret < 0) break;
std::vector<uint8_t> resampledData = resamplerDecoder->convert(frameDecode->data, frameDecode->nb_samples, frameDecode->channels, AV_SAMPLE_FMT_S16);
if (!resampledData.size()) continue;
std::copy(&resampledData.data()[0], &resampledData.data()[resampledData.size()], std::back_inserter(decodedData));
}
}
}
UPD 30.05.2020
I decided to refuse to use FFmpeg in my project and use libopus 1.3.1 instead, so I made a wrapper around it and it works fine.
I'm trying to transform a live stream with g726 and h264 to mp4. I decode g726 to pcm then use faac to encode pcm to aac. Every g726 audio packet I receive is 320 bytes. After decoding, the pcm size is 1280 bytes, so the sample number is 640. But the inputSamples which faacEncOpen gives me is 1024, and my inputFormat is FAAC_INPUT_16BIT. When I pass 640 to faacEncEncode, the sound is not good at all. Does anyone know how to fix this. Thanks in advance!
// (1) Open FAAC engine
hEncoder = faacEncOpen(nSampleRate, nChannels, &nInputSamples, &nMaxOutputBytes); // nInputSamples the function returns is 1024
if(hEncoder == NULL)
{
printf("[ERROR] Failed to call faacEncOpen()\n");
return -1;
}
nInputSamples = 640;// here overwrites the input samples returned from faacEncOpen
nPCMBufferSize = nInputSamples * nPCMBitSize / 8; // nPCMBitSize is 16
pbPCMBuffer = new BYTE [nPCMBufferSize];
pbAACBuffer = new BYTE [nMaxOutputBytes];
// (2.1) Get current encoding configuration
pConfiguration = faacEncGetCurrentConfiguration(hEncoder);
pConfiguration->inputFormat = FAAC_INPUT_16BIT;
// (2.2) Set encoding configuration
nRet = faacEncSetConfiguration(hEncoder, pConfiguration);
for(int i = 0; 1; i++)
{
nBytesRead = fread(pbPCMBuffer, 1, nPCMBufferSize, fpIn);
nInputSamples = nBytesRead * 8 / nPCMBitSize;
// (3) Encode
nRet = faacEncEncode(
hEncoder, (int*) pbPCMBuffer, nInputSamples, pbAACBuffer, nMaxOutputBytes);
fwrite(pbAACBuffer, 1, nRet, fpOut);
printf("%d: faacEncEncode returns %d\n", i, nRet);
if(nBytesRead <= 0)
{
break;
}
}
EDIT: I've updated the code below to resemble the progress I have made. I'm trying to write the .wav header myself. The code does not work properly as of now, the audio is not being written to the file properly. The code does not contain any attempts to convert it to a .flac file yet.
I am using a Raspberry Pi (Debian Linux) to record audio with the ALSA library. The recording works fine, but I need to encode the input audio into the FLAC codec.
This is where I get lost. I have spent a considerable amount of time trying to figure out how to convert this raw data into FLAC, but I keep coming up with examples of how to convert .wav files into .flac files.
Here is the current (updated) code I have for recording audio with ALSA (it may be a bit rough, I'm still picking up C++):
// Use the newer ALSA API
#define ALSA_PCM_NEW_HW_PARAMS_API
#include <alsa/asoundlib.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
struct Riff
{
char chunkId[4]; // "RIFF" (assuming char is 8 bits)
int chunkSize; // (assuming int is 32 bits)
char format[4]; // "WAVE"
};
struct Format
{
char chunkId[4]; // "fmt "
int chunkSize;
short format; // assuming short is 16 bits
short numChannels;
int sampleRate;
int byteRate;
short align;
short bitsPerSample;
};
struct Data
{
char chunkId[4]; // "data"
int chunkSize; // length of data
char* data;
};
struct Wave // Actual structure of a PCM WAVE file
{
Riff riffHeader;
Format formatHeader;
Data dataHeader;
};
int main(int argc, char *argv[])
{
void saveWaveFile(struct Wave *waveFile);
long loops;
int rc;
int size;
snd_pcm_t *handle;
snd_pcm_hw_params_t *params;
unsigned int sampleRate = 44100;
int dir;
snd_pcm_uframes_t frames;
char *buffer;
char *device = (char*) "plughw:1,0";
//char *device = (char*) "default";
printf("Capture device is %s\n", device);
/* Open PCM device for recording (capture). */
rc = snd_pcm_open(&handle, device, SND_PCM_STREAM_CAPTURE, 0);
if (rc < 0)
{
fprintf(stderr, "Unable to open PCM device: %s\n", snd_strerror(rc));
exit(1);
}
/* Allocate a hardware parameters object. */
snd_pcm_hw_params_alloca(¶ms);
/* Fill it in with default values. */
snd_pcm_hw_params_any(handle, params);
/* Set the desired hardware parameters. */
/* Interleaved mode */
snd_pcm_hw_params_set_access(handle, params, SND_PCM_ACCESS_RW_INTERLEAVED);
/* Signed 16-bit little-endian format */
snd_pcm_hw_params_set_format(handle, params, SND_PCM_FORMAT_S16_LE);
/* Two channels (stereo) */
snd_pcm_hw_params_set_channels(handle, params, 2);
/* 44100 bits/second sampling rate (CD quality) */
snd_pcm_hw_params_set_rate_near(handle, params, &sampleRate, &dir);
/* Set period size to 32 frames. */
frames = 32;
snd_pcm_hw_params_set_period_size_near(handle, params, &frames, &dir);
/* Write the parameters to the driver */
rc = snd_pcm_hw_params(handle, params);
if (rc < 0)
{
fprintf(stderr, "Unable to set HW parameters: %s\n", snd_strerror(rc));
exit(1);
}
/* Use a buffer large enough to hold one period */
snd_pcm_hw_params_get_period_size(params, &frames, &dir);
size = frames * 4; /* 2 bytes/sample, 2 channels */
buffer = (char *) malloc(size);
/* We want to loop for 5 seconds */
snd_pcm_hw_params_get_period_time(params, &sampleRate, &dir);
loops = 5000000 / sampleRate;
while (loops > 0)
{
loops--;
rc = snd_pcm_readi(handle, buffer, frames);
if (rc == -EPIPE)
{
/* EPIPE means overrun */
fprintf(stderr, "Overrun occurred.\n");
snd_pcm_prepare(handle);
} else if (rc < 0)
{
fprintf(stderr, "Error from read: %s\n", snd_strerror(rc));
} else if (rc != (int)frames)
{
fprintf(stderr, "Short read, read %d frames.\n", rc);
}
if (rc != size) fprintf(stderr, "Short write: wrote %d bytes.\n", rc);
}
Wave wave;
strcpy(wave.riffHeader.chunkId, "RIFF");
wave.riffHeader.chunkSize = 36 + size;
strcpy(wave.riffHeader.format, "WAVE");
strcpy(wave.formatHeader.chunkId, "fmt");
wave.formatHeader.chunkSize = 16;
wave.formatHeader.format = 1; // PCM, other value indicates compression
wave.formatHeader.numChannels = 2; // Stereo
wave.formatHeader.sampleRate = sampleRate;
wave.formatHeader.byteRate = sampleRate * 2 * 2;
wave.formatHeader.align = 2 * 2;
wave.formatHeader.bitsPerSample = 16;
strcpy(wave.dataHeader.chunkId, "data");
wave.dataHeader.chunkSize = size;
wave.dataHeader.data = buffer;
saveWaveFile(&wave);
snd_pcm_drain(handle);
snd_pcm_close(handle);
free(buffer);
return 0;
}
void saveWaveFile(struct Wave *waveFile)
{
FILE *file = fopen("test.wav", "wb");
size_t written;
if (file == NULL)
{
fprintf(stderr, "Cannot open file for writing.\n");
exit(1);
}
written = fwrite(waveFile, sizeof waveFile[0], 1, file);
fclose(file);
if (written < 1);
{
fprintf(stderr, "Writing to file failed, error %d.\n", written);
exit(1);
}
}
How would I go about converting the PCM data into the FLAC and save it to disk for later use? I have downloaded libflac-dev already and just need an example to go off of.
The way I am doing it right now:
./capture > test.raw // or ./capture > test.flac
The way it should be (program does everything for me):
./capture
If I understand the FLAC::Encoder::File documentation, you can do something like
#include <FLAC++/encoder.h>
FLAC::Encoder::File encoder;
encoder.init("outfile.flac");
encoder.process(buffer, samples);
encoder.finish();
where buffer is an array (of size samples) of 32-bit integer pointers.
Unfortunately, I know next to nothing about audio encoding so I can't speak for any other options. Good luck!
Please refer to the below code :
FLAC Encoder Test Code
This example is using a wav file as an input and then encodes it into FLAC.
As I understand, there is no major difference b/w WAV file and your RAW data, I think you can modify this code to directly read the "buffer" and convert it. You already have all the related information (Channel/Bitrate etc) so it should not be much of a problem to remove the WAV header reading code.
Please note: this is a modified version of the Flac Encoder sample from their git repo.
It includes some comments and hints on how to change it to OP's requirements, entire source for this will be a little bit long.
And do note that this is the C API, which tends to be a bit more complex than the C++ one. But it is fairly easy to convert between the two once you get the idea.
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include "share/compat.h"
#include "FLAC/metadata.h"
#include "FLAC/stream_encoder.h"
/* this call back is what tells your program the progress that the encoder has made */
static void progress_callback(const FLAC__StreamEncoder *encoder, FLAC__uint64 bytes_written, FLAC__uint64 samples_written, unsigned frames_written, unsigned total_frames_estimate, void *client_data);
#define READSIZE 1024
static unsigned total_samples = 0; /* can use a 32-bit number due to WAVE size limitations */
/* buffer is where we record to, in your case what ALSA writes to */
/* Note the calculation here to take the total bytes that the buffer takes */
static FLAC__byte buffer[READSIZE/*samples*/ * 2/*bytes_per_sample*/ * 2/*channels*/];
/* pcm is input to FLAC encoder */
/* the PCM data should be here, bps is 4 here...but we are allocating ints! */
static FLAC__int32 pcm[READSIZE/*samples*/ * 2/*channels*/];
int main(int argc, char *argv[])
{
FLAC__bool ok = true;
FLAC__StreamEncoder *encoder = 0;
FLAC__StreamEncoderInitStatus init_status;
FLAC__StreamMetadata *metadata[2];
FLAC__StreamMetadata_VorbisComment_Entry entry;
FILE *fin;
unsigned sample_rate = 0;
unsigned channels = 0;
unsigned bps = 0;
if((fin = fopen(argv[1], "rb")) == NULL) {
fprintf(stderr, "ERROR: opening %s for output\n", argv[1]);
return 1;
}
/* set sample rate, bps, total samples to encode here, these are dummy values */
sample_rate = 44100;
channels = 2;
bps = 16;
total_samples = 5000;
/* allocate the encoder */
if((encoder = FLAC__stream_encoder_new()) == NULL) {
fprintf(stderr, "ERROR: allocating encoder\n");
fclose(fin);
return 1;
}
ok &= FLAC__stream_encoder_set_verify(encoder, true);
ok &= FLAC__stream_encoder_set_compression_level(encoder, 5);
ok &= FLAC__stream_encoder_set_channels(encoder, channels);
ok &= FLAC__stream_encoder_set_bits_per_sample(encoder, bps);
ok &= FLAC__stream_encoder_set_sample_rate(encoder, sample_rate);
ok &= FLAC__stream_encoder_set_total_samples_estimate(encoder, total_samples);
/* sample adds meta data here I've removed it for clarity */
/* initialize encoder */
if(ok) {
/* client data is whats the progress_callback is called with, any objects you need to update on callback can be passed thru this pointer */
init_status = FLAC__stream_encoder_init_file(encoder, argv[2], progress_callback, /*client_data=*/NULL);
if(init_status != FLAC__STREAM_ENCODER_INIT_STATUS_OK) {
fprintf(stderr, "ERROR: initializing encoder: %s\n", FLAC__StreamEncoderInitStatusString[init_status]);
ok = false;
}
}
/* read blocks of samples from WAVE file and feed to encoder */
if(ok) {
size_t left = (size_t)total_samples;
while(ok && left) {
/* record using ALSA and set SAMPLES_IN_BUFFER */
/* convert the packed little-endian 16-bit PCM samples from WAVE into an interleaved FLAC__int32 buffer for libFLAC */
/* why? because bps=2 means that we are dealing with short int(16 bit) samples these are usually signed if you do not explicitly say that they are unsigned */
size_t i;
for(i = 0; i < SAMPLES_IN_BUFFER*channels; i++) {
/* THIS. this isn't the only way to convert between formats, I do not condone this because at first the glance the code seems like it's processing two channels here, but it's not it's just copying 16bit data to an int array, I prefer to use proper type casting, none the less this works so... */
pcm[i] = (FLAC__int32)(((FLAC__int16)(FLAC__int8)buffer[2*i+1] << 8) | (FLAC__int16)buffer[2*i]);
}
/* feed samples to encoder */
ok = FLAC__stream_encoder_process_interleaved(encoder, pcm, SAMPLES_IN_BUFFER);
left-=SAMPLES_IN_BUFFER;
}
}
ok &= FLAC__stream_encoder_finish(encoder);
fprintf(stderr, "encoding: %s\n", ok? "succeeded" : "FAILED");
fprintf(stderr, " state: %s\n", FLAC__StreamEncoderStateString[FLAC__stream_encoder_get_state(encoder)]);
FLAC__stream_encoder_delete(encoder);
fclose(fin);
return 0;
}
/* the updates from FLAC's encoder system comes here */
void progress_callback(const FLAC__StreamEncoder *encoder, FLAC__uint64 bytes_written, FLAC__uint64 samples_written, unsigned frames_written, unsigned total_frames_estimate, void *client_data)
{
(void)encoder, (void)client_data;
fprintf(stderr, "wrote %" PRIu64 " bytes, %" PRIu64 "/%u samples, %u/%u frames\n", bytes_written, samples_written, total_samples, frames_written, total_frames_estimate);
}