Android回音噪音处理Demo

这是一个开源项目，核心来自webrtc项目。

据网友测试：大多数手机上比较完美，没啥回声，但在小米上比较明显噪音。

将实现各个平台上能快速使用的音频处理库。
核心算法包括：
NS（Noise Suppression 噪声抑制）
VAD（Voice Activity Detection 静音检测）
AECM（Acoustic Echo Canceller for Mobile 声学回声消除）
AGC（Auto Gain Control 自动增益控制）现在只有一个Android Demo 。
核心代码如下：

package net.iwebrtc.audioprocess.sdk;import android.util.Log;public class AudioProcess {private static final String TAG = "AudioProcess";static {String[] LIBS = new String[] { "audio_process" };for (int i = 0; i < LIBS.length; i++) {try {System.loadLibrary(LIBS[i]);} catch (UnsatisfiedLinkError e) {Log.e(TAG, "Couldn't load lib: " + LIBS[i] + " - " + e.getMessage());}}}private long nativeAudioProcess;public AudioProcess() {nativeAudioProcess = create();}private native long create();public native boolean init(int sample_rate, int number_bytes_per_sample, int channels);public int calculateBufferSize(int sample_rate, int number_bytes_per_sample, int channels) {return sample_rate * channels * number_bytes_per_sample / 100;}public native boolean processStream10msData(byte[] data, int length, byte[] out);public native boolean AnalyzeReverseStream10msData(byte[] data, int length);public native boolean destroy();}

地址：https://github.com/xiaomo/AudioProcess

演示Demo:http://download.csdn.net/detail/moming_2013/8209547

刚刚开始做，肯定还有不少bug，会不断完善。

音频处理模块(APM)使用翻译：

The Audio Processing Module (APM) provides a collection of voice processing components designed for real-time communications software.
音频处理模块(APM)提供了一个音频处理组件的集合，这些组件为实时通讯软件而设计。

APM operates on two audio streams on a frame-by-frame basis.
APM操作两个一帧接一帧的音频流。一个主流，一个逆向流。

Frames of the primary stream, on which all processing is applied, are passed to |ProcessStream()|.
所有的处理都将被应用到主流的帧  ，这些帧通过ProcessStream函数传递过去。

Frames of the reverse direction stream, which are used for analysis by some components, are passed to |AnalyzeReverseStream()|.
逆向流的帧被一些组件拿去分析，逆向流的帧通过AnalyzeReverseStream函数传递过去。

On the client-side, this will typically be the near-end (capture) and far-end (render) streams, respectively.
在客户端，这是典型的相互独立的近端流和远端流。

APM should be placed in the signal chain as close to the audio hardware abstraction layer (HAL) as possible.
APM在信号链中应该尽快能被放置在接近硬件抽象层的地方。

On the server-side, the reverse stream will normally not be used, with processing occurring on each incoming stream.
在服务端，逆向流通常不被使用。

Component interfaces follow a similar pattern and are accessed through corresponding getters in APM.
组件接口遵循类似的模式设计，通过APM中相应的getter函数访问。

All components are disabled at create-time, with default settings that are recommended for most situations.
所有的组件在创建时都是不可用的，在大多数情况下，推荐这种默认的设置。

New settings can be applied without enabling a component.
使一个组件可用，新的设置将会被应用到APM。
Enabling a component triggers memory allocation and initialization to allow it to start processing the streams.
使一个组件可用会触发内存分配和初始化，以允许这个组件开始处理音频流。

Thread safety is provided with the following assumptions to reduce locking overhead:
基于下面的假设，提供了线程安全，用以减少锁的开销：
   1. The stream getters and setters are called from the same thread as ProcessStream(). More precisely, stream functions are never called  concurrently with ProcessStream().
     在ProcessStream的时候，流的getter和setters方法应该在同一个线程被调用。更确切的说，流的方法绝对不能在ProcessStream的同时被调用。
   2. Parameter getters are never called concurrently with the corresponding setter.
    参数的getter和setter绝不能同时条用。

APM accepts only 16-bit linear PCM audio data in frames of 10 ms. Multiple channels should be interleaved.
APM只接受10 ms帧的16位线性PCM音频数据。多声道应该是交错。

Usage example, omitting error checking:
使用示例，省略了错误检查：
AudioProcessing* apm = AudioProcessing::Create(0);
1 创建一个APM
apm->set_sample_rate_hz(32000); // Super-wideband processing.
2 设置采样率
// Mono capture and stereo render.
录音单声道，播放多声道。
apm->set_num_channels(1, 1);
3 设置主流处理的输入和输出声道数
apm->set_num_reverse_channels(2);
4 设置逆向流的声道数

apm->high_pass_filter()->Enable(true);
5 让快速通过过滤组件可用

apm->echo_cancellation()->enable_drift_compensation(false);
apm->echo_cancellation()->Enable(true);
6 设置ACE组件

apm->noise_reduction()->set_level(kHighSuppression);
apm->noise_reduction()->Enable(true);
7 设置噪音消除组件

apm->gain_control()->set_analog_level_limits(0, 255);
apm->gain_control()->set_mode(kAdaptiveAnalog);
apm->gain_control()->Enable(true);
8 设置自动增益组件

apm->voice_detection()->Enable(true);
9 设置静音检测组件
// Start a voice call...
// ... Render frame arrives bound for the audio HAL ...
apm->AnalyzeReverseStream(render_frame);
10 开始语音通话，让设备播放收到了的帧，把帧通过逆向流传给APM

// ... Capture frame arrives from the audio HAL ...
// Call required set_stream_ functions.
apm->set_stream_delay_ms(delay_ms);
apm->gain_control()->set_stream_analog_level(analog_level);

apm->ProcessStream(capture_frame);
11 从设备采集帧，然后必须调用设置流的方法set_stream_delay_ms和set_stream_analog_level,然后处理采集的帧。

// Call required stream_ functions.
analog_level = apm->gain_control()->stream_analog_level();
has_voice = apm->stream_has_voice();
12 然后调动流的方法stream_analog_level和stream_has_voice是有必要的

// Repeate render and capture processing for the duration of the call...
// Start a new call...
apm->Initialize();
13 开始一个新的会话，然后调用Initialize。

// Close the application...
delete apm;
14 结束app，删除amp

更多相关文章

随机推荐