This series of articles Master link: Thematic sub-directory Android Framework Class Audio Subsystem
Summary and description of key points in this chapter:
This chapter mainly focuses on ➕ the upper left recording part of the above mind map. It mainly explains the basic knowledge of some recordings, and at the same time has an understanding of the recording process of android through a Native test program.
1 Basic knowledge of recording
@ 1 Relationship between PCM and WAV
PCM audio data is the original audio data, which cannot be played by the player. You need to add a header to it to indicate how many channels the sound has, what the sampling rate is, and so on. Convert
PCM audio data to WAV format so that other players can play it.
@ 2 3 key parameters of recording
- Sampling rate: the number of times the sound wave is sampled per second. Common sampling rates are 8000, 11025, 44100. . .
- Sampling accuracy: Currently fixed to 16bit on Android system.
- Number of channels: two channels (Stereo stereo, each sample point records the value of the left and right channels) or mono (Mono).
2 Recording test procedure
There are three key documents involved:
- AudioRecordTest.cpp: recording program, finally output pcm audio format data.
- pcm2wav.cpp: add header information in pcm format and convert to WAV format.
- Android.mk: Compile configuration.
@ 1 AudioRecordTest.cpp, used to record pcm data, but no header information is added here. code show as below:
#include <utils/Log.h>
#include <media/AudioRecord.h>
#include <stdlib.h>
using namespace android;
//==============================================
// Audio Record Defination
//==============================================
#ifdef LOG_TAG
#undef LOG_TAG
#endif
#define LOG_TAG "AudioRecordTest"
static pthread_t g_AudioRecordThread;
static pthread_t * g_AudioRecordThreadPtr = NULL;
volatile bool g_bQuitAudioRecordThread = false;
volatile int g_iInSampleTime = 0;
int g_iNotificationPeriodInFrames = 8000/10;
// g_iNotificationPeriodInFrames should be change when sample rate changes.
static void * AudioRecordThread(int sample_rate, int channels, void *fileName)
{
uint64_t inHostTime = 0;
void * inBuffer = NULL;
audio_source_t inputSource = AUDIO_SOURCE_MIC;
audio_format_t audioFormat = AUDIO_FORMAT_PCM_16_BIT;
audio_channel_mask_t channelConfig = AUDIO_CHANNEL_IN_MONO;
int bufferSizeInBytes;
int sampleRateInHz = sample_rate; //8000; //44100;
android::AudioRecord * pAudioRecord = NULL;
FILE * g_pAudioRecordFile = NULL;
char * strAudioFile = (char *)fileName;
int iNbChannels = channels; // 1 channel for mono, 2 channel for streo
int iBytesPerSample = 2; // 16bits pcm, 2Bytes
int frameSize = 0; // frameSize = iNbChannels * iBytesPerSample
size_t minFrameCount = 0; // get from AudroRecord object
int iWriteDataCount = 0; // how many data are there write to file
// log the thread id for debug info
ALOGD("%s Thread ID = %d \n", __FUNCTION__, pthread_self());
g_iInSampleTime = 0;
g_pAudioRecordFile = fopen(strAudioFile, "wb+");
//printf("sample_rate = %d, channels = %d, iNbChannels = %d, channelConfig = 0x%x\n", sample_rate, channels, iNbChannels, channelConfig);
//iNbChannels = (channelConfig == AUDIO_CHANNEL_IN_STEREO) ? 2 : 1;
if (iNbChannels == 2) {
channelConfig = AUDIO_CHANNEL_IN_STEREO;
}
printf("sample_rate = %d, channels = %d, iNbChannels = %d, channelConfig = 0x%x\n", sample_rate, channels, iNbChannels, channelConfig);
frameSize = iNbChannels * iBytesPerSample;
android::status_t status = android::AudioRecord::getMinFrameCount(
&minFrameCount, sampleRateInHz, audioFormat, channelConfig);
if(status != android::NO_ERROR)
{
ALOGE("%s AudioRecord.getMinFrameCount fail \n", __FUNCTION__);
goto exit ;
}
ALOGE("sampleRateInHz = %d minFrameCount = %d iNbChannels = %d channelConfig = 0x%x frameSize = %d ",
sampleRateInHz, minFrameCount, iNbChannels, channelConfig, frameSize);
bufferSizeInBytes = minFrameCount * frameSize;
//申请内存
inBuffer = malloc(bufferSizeInBytes);
if(inBuffer == NULL)
{
ALOGE("%s alloc mem failed \n", __FUNCTION__);
goto exit ;
}
g_iNotificationPeriodInFrames = sampleRateInHz/10;
//创建AudioRecord
pAudioRecord = new android::AudioRecord();
if(NULL == pAudioRecord)
{
ALOGE(" create native AudioRecord failed! ");
goto exit;
}
//set操作
pAudioRecord->set( inputSource,
sampleRateInHz,
audioFormat,
channelConfig,
0,
NULL, //AudioRecordCallback,
NULL,
0,
true,
0);
if(pAudioRecord->initCheck() != android::NO_ERROR)
{
ALOGE("AudioTrack initCheck error!");
goto exit;
}
//开始录音
if(pAudioRecord->start()!= android::NO_ERROR)
{
ALOGE("AudioTrack start error!");
goto exit;
}
while (!g_bQuitAudioRecordThread)
{
//从底层读取音频数据
int readLen = pAudioRecord->read(inBuffer, bufferSizeInBytes);
int writeResult = -1;
if(readLen > 0)
{
iWriteDataCount += readLen;
if(NULL != g_pAudioRecordFile)
{
//将音频数据写入指定文件中
writeResult = fwrite(inBuffer, 1, readLen, g_pAudioRecordFile);
if(writeResult < readLen)
{
ALOGE("Write Audio Record Stream error");
}
}
//ALOGD("readLen = %d writeResult = %d iWriteDataCount = %d", readLen, writeResult, iWriteDataCount);
}
else
{
ALOGE("pAudioRecord->read readLen = 0");
}
}
exit:
if(NULL != g_pAudioRecordFile)
{
fflush(g_pAudioRecordFile);
fclose(g_pAudioRecordFile);
g_pAudioRecordFile = NULL;
}
if(pAudioRecord)
{
pAudioRecord->stop();
//delete pAudioRecord;
//pAudioRecord == NULL;
}
if(inBuffer)
{
free(inBuffer);
inBuffer = NULL;
}
ALOGD("%s Thread ID = %d quit\n", __FUNCTION__, pthread_self());
return NULL;
}
int main(int argc, char **argv)
{
if (argc != 4)
{
printf("Usage:\n");
printf("%s <sample_rate> <channels> <out_file>\n", argv[0]);
return -1;
}
AudioRecordThread(strtol(argv[1], NULL, 0), strtol(argv[2], NULL, 0), argv[3]);
return 0;
}@ 2 pcm2wav.cpp
#include <stdio.h>
#include <string.h>
#include <stdlib.h>
/* 参考https://blog.csdn.net/u010011236/article/details/53026127 */
/**
* Convert PCM16LE raw data to WAVE format
* @param pcmpath Input PCM file.
* @param channels Channel number of PCM file.
* @param sample_rate Sample rate of PCM file.
* @param wavepath Output WAVE file.
*/
int simplest_pcm16le_to_wave(const char *pcmpath, int sample_rate, int channels, const char *wavepath)
{
typedef struct WAVE_HEADER{
char fccID[4]; //内容为""RIFF
unsigned long dwSize; //最后填写，WAVE格式音频的大小
char fccType[4]; //内容为"WAVE"
}WAVE_HEADER;
typedef struct WAVE_FMT{
char fccID[4]; //内容为"fmt "
unsigned long dwSize; //内容为WAVE_FMT占的字节数，为16
unsigned short wFormatTag; //如果为PCM，改值为 1
unsigned short wChannels; //通道数，单通道=1，双通道=2
unsigned long dwSamplesPerSec;//采用频率
unsigned long dwAvgBytesPerSec;/* ==dwSamplesPerSec*wChannels*uiBitsPerSample/8 */
unsigned short wBlockAlign;//==wChannels*uiBitsPerSample/8
unsigned short uiBitsPerSample;//每个采样点的bit数，8bits=8, 16bits=16
}WAVE_FMT;
typedef struct WAVE_DATA{
char fccID[4]; //内容为"data"
unsigned long dwSize; //==NumSamples*wChannels*uiBitsPerSample/8
}WAVE_DATA;
int bits = 16;
WAVE_HEADER pcmHEADER;
WAVE_FMT pcmFMT;
WAVE_DATA pcmDATA;
unsigned short m_pcmData;
FILE *fp, *fpout;
fp = fopen(pcmpath, "rb+");
if(fp==NULL)
{
printf("Open pcm file error.\n");
return -1;
}
fpout = fopen(wavepath, "wb+");
if(fpout==NULL)
{
printf("Create wav file error.\n");
return -1;
}
/* WAVE_HEADER */
memcpy(pcmHEADER.fccID, "RIFF", strlen("RIFF"));
memcpy(pcmHEADER.fccType, "WAVE", strlen("WAVE"));
fseek(fpout, sizeof(WAVE_HEADER), 1); //1=SEEK_CUR
/* WAVE_FMT */
memcpy(pcmFMT.fccID, "fmt ", strlen("fmt "));
pcmFMT.dwSize = 16;
pcmFMT.wFormatTag = 1;
pcmFMT.wChannels = channels;
pcmFMT.dwSamplesPerSec = sample_rate;
pcmFMT.uiBitsPerSample = bits;
/* ==dwSamplesPerSec*wChannels*uiBitsPerSample/8 */
pcmFMT.dwAvgBytesPerSec = pcmFMT.dwSamplesPerSec*pcmFMT.wChannels*pcmFMT.uiBitsPerSample/8;
/* ==wChannels*uiBitsPerSample/8 */
pcmFMT.wBlockAlign = pcmFMT.wChannels*pcmFMT.uiBitsPerSample/8;
fwrite(&pcmFMT, sizeof(WAVE_FMT), 1, fpout);
/* WAVE_DATA */
memcpy(pcmDATA.fccID, "data", strlen("data"));
pcmDATA.dwSize = 0;
fseek(fpout, sizeof(WAVE_DATA), SEEK_CUR);
fread(&m_pcmData, sizeof(unsigned short), 1, fp);
while(!feof(fp))
{
pcmDATA.dwSize += 2;
fwrite(&m_pcmData, sizeof(unsigned short), 1, fpout);
fread(&m_pcmData, sizeof(unsigned short), 1, fp);
}
/*pcmHEADER.dwSize = 44 + pcmDATA.dwSize;*/
pcmHEADER.dwSize = 36 + pcmDATA.dwSize;
rewind(fpout);
fwrite(&pcmHEADER, sizeof(WAVE_HEADER), 1, fpout);
fseek(fpout, sizeof(WAVE_FMT), SEEK_CUR);
fwrite(&pcmDATA, sizeof(WAVE_DATA), 1, fpout);
fclose(fp);
fclose(fpout);
return 0;
}
int main(int argc, char **argv)
{
if (argc != 5)
{
printf("Usage:\n");
printf("%s <input pcm file> <sample_rate> <channels> <output wav file>\n", argv[0]);
return -1;
}
simplest_pcm16le_to_wave(argv[1], strtol(argv[2], NULL, 0), strtol(argv[3], NULL, 0), argv[4]);
return 0;
}@ 3 Android.mk implementation
LOCAL_PATH:= $(call my-dir)
include $(CLEAR_VARS)
LOCAL_SRC_FILES:= \
AudioRecordTest.cpp
LOCAL_SHARED_LIBRARIES := \
libcutils \
libutils \
libmedia
LOCAL_MODULE:= AudioRecordTest
LOCAL_MODULE_TAGS := tests
include $(BUILD_EXECUTABLE)
include $(CLEAR_VARS)
LOCAL_SRC_FILES:= \
pcm2wav.cpp
LOCAL_SHARED_LIBRARIES := \
libcutils \
libutils \
libmedia
LOCAL_MODULE:= pcm2wav
LOCAL_MODULE_TAGS := tests
include $(BUILD_EXECUTABLE)3 Test procedure execution steps
@ 1 Start recording:
./AudioRecordTest 44100 2 my.pcm@ 2 File format conversion (PCM to WAV):
./pcm2wav my.pcm 44100 2 my.wav4 Encountered problems and analysis
@ 1 tinyplay can't play mono sound
If you must play, you can only use other players.
@ 2 Why do I use dual sound when recording, and only one ear has sound during playback? And when using monophonic recording, both ears have sound during playback?
- The hardware and driver are dual-channel; but we only received one MIC, so in the dual-channel data obtained by the driver during recording, one of the channel data is always 0, and if the AudioRecordTest recording is specified for dual-channel Channel, then one of the channels in the obtained PCM data is always 0, and it will cause only one ear to have sound when it is played.
- If Mono is specified during AudioRecordTest recording, the obtained PCM data contains only one channel data, which is a mix of hardware left and right channels. This mix is realized by AudioFlinger system. When playing mono data during playback, The AudioFlinger system will send mono data to both the hardware Left DAC (left channel) and the hardware Right DAC (right channel), so both ears can hear the sound
