android 采集PCM音频数据并播放(支持USB摄像头MIC)

84 阅读 0 评论 56 点赞

我是靠谱客的博主斯文砖头，最近开发中收集的这篇文章主要介绍android 采集PCM音频数据并播放(支持USB摄像头MIC)，觉得挺不错的，现在分享给大家，希望可以做个参考。

概述

平台

RK3288 + Android 7.1

开发

在开始之前需要先了解下当前的音频设备情况:

查看当前支持的声卡设备(找到card[0, N])

rk3288:/proc/asound # ll
total 0
lrwxrwxrwx 1 root root 5 2020-04-15 13:51 Camera -> card1
dr-xr-xr-x 5 root root 0 2020-04-15 13:51 card0
dr-xr-xr-x 3 root root 0 2020-04-15 13:51 card1
-r--r--r-- 1 root root 0 2020-04-15 13:51 cards
-r--r--r-- 1 root root 0 2020-04-15 13:51 devices
-r--r--r-- 1 root root 0 2020-04-15 13:51 hwdep
-r--r--r-- 1 root root 0 2020-04-15 13:51 pcm
lrwxrwxrwx 1 root root 5 2020-04-15 13:51 rockchipes8316c -> card0
-r--r--r-- 1 root root 0 2020-04-15 13:51 timers
-r--r--r-- 1 root root 0 2020-04-15 13:51 version

查看cards中的内容:

rk3288:/proc/asound # cat cards
0 [rockchipes8316c]: rockchip_es8316 - rockchip,es8316-codec
rockchip,es8316-codec
1 [Camera
]: USB-Audio - USB Camera
Generic USB Camera at usb-ff540000.usb-1.2, high speed

当前使用的声卡信息

rk3288:/proc/asound # tinypcminfo -D 0
Info for card 0, device 0:
PCM out:
Access:	0x000009
Format[0]:	0x000044
Format[1]:	00000000
Format Name:	S16_LE, S24_LE
Subformat:	0x000001
Rate:	min=8000Hz	max=96000Hz
Channels:	min=2
max=2
Sample bits:	min=16
max=32
Period size:	min=32
max=65536
Period count:	min=2
max=4096
PCM in:
Access:	0x000009
Format[0]:	0x000044
Format[1]:	00000000
Format Name:	S16_LE, S24_LE
Subformat:	0x000001
Rate:	min=8000Hz	max=96000Hz
Channels:	min=2
max=2
Sample bits:	min=16
max=32
Period size:	min=32
max=65536
Period count:	min=2
max=4096

采集API:AudioRecord


//audioSource: MediaRecorder.AudioSource.CAMCORDER, MediaRecorder.AudioSource.MIC 等
//
一般情况下会用到这两个
//sampleRateInHz: 常用的有: 8000,11025,16000,22050,44100,96000;
//channelConfig: AudioFormat.CHANNEL_CONFIGURATION_DEFAULT, 单声道和立声
//audioFormat: AudioFormat.ENCODING_PCM_16BIT, 测试中要捕获PCM数据, 其它参数未研究.
//bufferSizeInBytes: 由AudioRecord.getMinBufferSize(sampleRateInHz, channelConfig, audioFormat)获得
public AudioRecord(int audioSource, int sampleRateInHz, int channelConfig, int audioFormat,
int bufferSizeInBytes)

播放API: AudioTrack


//streamType: AudioManager.STREAM_MUSIC, 输出通道, 也可以选择ALARM等
//sampleRateInHz: 常用的有: 8000,11025,16000,22050,44100,96000;
//channelConfig: AudioFormat.CHANNEL_CONFIGURATION_DEFAULT, 单声道和立声
//audioFormat: AudioFormat.ENCODING_PCM_16BIT, 测试中要捕获PCM数据, 其它参数未研究.
//bufferSizeInBytes: 
//mode: AudioTrack中有MODE_STATIC和MODE_STREAM两种分类。
//
STREAM的意思是由用户在应用程序通过write方式把数据一次一次得写到audiotrack中。
//
这个和我们在socket中发送数据一样，应用层从某个地方获取数据，例如通过编解码得到PCM数据，然后write到audiotrack。
//
这种方式的坏处就是总是在JAVA层和Native层交互，效率损失较大。
//
而STATIC的意思是一开始创建的时候，就把音频数据放到一个固定的buffer，然后直接传给audiotrack，
//
后续就不用一次次得write了。AudioTrack会自己播放这个buffer中的数据。
//
这种方法对于铃声等内存占用较小，延时要求较高的声音来说很适用。
public AudioTrack(int streamType, int sampleRateInHz, int channelConfig, int audioFormat,
int bufferSizeInBytes, int mode)

边采集边播放PCM代码(仅核心代码, 不完整)


//Audio Parameter
static class AudioParam{
static final int FREQ_8000 = 8000;
static final int FREQ_11025 = 11025;
static final int FREQ_16000 = 16000;
static final int FREQ_22050 = 22050;
static final int FREQ_44100 = 44100;
static final int FREQ_96000 = 96000;
int device;
int channel;
int bitFormat;
int buffSize;
//int doubleBuffSize;
int buffSizeOut;
int rate;
public AudioParam(int device, int rate,
int channel, int bitFormat){
this.device = device;
this.rate = rate;
this.channel = channel;
this.bitFormat = bitFormat;
initBufferSize();
}
private void initBufferSize() {
buffSize = AudioRecord.getMinBufferSize(rate, channel, bitFormat);
buffSizeOut = AudioTrack.getMinBufferSize(rate, channel, bitFormat);
Logger.d("AudioPCM", "buffSize(" + buffSize + "), buffSizeOut(" + buffSizeOut + ")");
if(buffSizeOut < 0)buffSizeOut = buffSize;
//doubleBuffSize = buffSize * 2;
}
static AudioParam getDefaultInParam(){
return new AudioParam(MediaRecorder.AudioSource.MIC,
FREQ_44100,
AudioFormat.CHANNEL_CONFIGURATION_DEFAULT,
AudioFormat.ENCODING_PCM_16BIT);
}
static AudioParam getDefaultOutParam(){
return new AudioParam(AudioManager.STREAM_MUSIC,
FREQ_44100,
AudioFormat.CHANNEL_CONFIGURATION_DEFAULT,
AudioFormat.ENCODING_PCM_16BIT);
}
}
//Thread for capture audio
class CaptureThread extends Thread{
@Override
public void run() {
AudioRecord mic = new AudioRecord(audioParamIn.device, audioParamIn.rate, audioParamIn.channel,
audioParamIn.bitFormat, audioParamIn.buffSize);
mic.startRecording();
byte[] pcmBuffer = new byte[2048];
while (!Thread.interrupted()) {
int size = mic.read(pcmBuffer, 0, pcmBuffer.length);
Logger.d(TAG, "read " + size + " bytes");
if (size <= 0) {
break;
}else{
if(playThd != null){
playThd.write(pcmBuffer, size);
}
}
}
mic.stop();
mic.release();
}
}
//Thread for play
class PlayThread{
private AudioTrack mAudioTrack;
PlayThread(){
try {
audioParamOut = new AudioParam(AudioManager.STREAM_MUSIC,
AudioParam.FREQ_44100,
AudioFormat.CHANNEL_CONFIGURATION_DEFAULT,
AudioFormat.ENCODING_PCM_16BIT);
createAudioTrack();
mAudioTrack.play();
} catch (Exception e) {
e.printStackTrace();
}
}
void write(byte[] bs, int size){
Logger.d(TAG, "write " + size + " bytes");
if(mAudioTrack != null){
mAudioTrack.write(bs, 0, size);
}
}
void stop(){
if(mAudioTrack != null){
mAudioTrack.stop();
}
}
private void createAudioTrack() throws Exception{
//
STREAM_ALARM：警告声
//
STREAM_MUSCI：音乐声，例如music等
//
STREAM_RING：铃声
//
STREAM_SYSTEM：系统声音
//
STREAM_VOCIE_CALL：电话声音
mAudioTrack = new AudioTrack(audioParamOut.device,
audioParamOut.rate,
audioParamOut.channel,
audioParamOut.bitFormat,
audioParamOut.buffSizeOut,
AudioTrack.MODE_STREAM);
}
}

扩展

需要确定能正常的识别到USB 摄像头的MIC输入设备(这个过程走了许多弯路)
在测试过程中发现 MediaRecorder.AudioSource.CAMCORDER, MediaRecorder.AudioSource.MIC对应的设备并不是固定的
- 当未接USB摄像头的时候 MediaRecorder.AudioSource.MIC 对应的是 主板的MIC
- 当接上USB摄像头后, MediaRecorder.AudioSource.MIC 对应的是USB摄像头上的MIC.