Abson / Camerartmpsdk

Which is CameraRtmpSDK

CameraRtmpSDK is an audio and video based on ffmpeg SDK, is committed to creating a cross-platform audio and video sampling, encoding, mixing, protocol upload program. But all current creation is based on the iOS platform and CameraRtmpSDK was coded by c++.

Blog to Introduction

利用FFmpeg 开发音视频流（三）——将视频 YUV 格式编码成 H264

深入浅出理解视频编码H264结构

i did write a blog to introduce this project , hope it can help you to comprehend it.

How to Use

If you want to test Vido (H.264编码)

self.session = [[ABSSimpleSession alloc] initWithVideoSize:CGSizeMake(640, 480)
                                                       fps:30
                                                   bitrate:1000000
                                   useInterfaceOrientation:true
                                               cameraState:ABSCameraStateFront previewFrame:self.view.bounds];
self.session.delegate = self;
[self.view addSubview:self.session.previewView];

[self.session startVideoRecord];

// close it 
[self.session endVidoeRecord];

if you want to test audio for AAC (AAC编码)

self.session = [[ABSSimpleSession alloc] initWithAudioSampleRate:16000 bitRate:16000
                                                      channelCount:2 encode:ABSEncodeTypeAAC];
self.session.delegate = self;
[self.session startAudioRecord];

// close it 
[self.session endAudioRecord];

if you want to test audio for Opus (Opus编码)

self.session = [[ABSSimpleSession alloc] initWithAudioSampleRate:16000 bitRate:16000
                                                      channelCount:2 encode:ABSEncodeTypeOpus];
self.session.delegate = self;
[self.session startAudioRecord];

// close it 
[self.session endAudioRecord];

if you want to test amix（混音） of ffmpeg

NSString* path1 = [[NSBundle mainBundle] pathForResource:@"audio1.wav" ofType:nil];
NSString* path2 = [[NSBundle mainBundle] pathForResource:@"audio2.wav" ofType:nil];
NSString* outputPath = [NSSearchPathForDirectoriesInDomains(NSDocumentDirectory, NSUserDomainMask, true).lastObject 
stringByAppendingPathComponent:@"audio127.wav"];

const char* input1 = [path1 cStringUsingEncoding:NSUTF8StringEncoding];
const char* input2 = [path2 cStringUsingEncoding:NSUTF8StringEncoding];
const char* output = [outputPath cStringUsingEncoding:NSUTF8StringEncoding];

std::vector<std::string> inputs{std::string(input1), std::string(input2)};

PushSDK::ffmpeg::audio_mixer mixer(inputs, output);
mixer.StartMixAudio();

Update

2017.11.1 Update project to runnable and and modify test API, enjoy it.

2018.1.2 Fix audio encode bug, make it encode correctly，enjoy it.

Feature

As far as this moment, This project looks very confusing，i will take more time make it more usable. hope you can supports me, leaving you stars, Thanks.