iPhone 16 Pro Max上AVFoundation音频压缩失败问题求助

在使用AVFoundation压缩iPhone 16 Pro Max录制的视频时，音频压缩环节出现异常。调用AVAssetWriterInput的appendSampleBuffer方法后，写入器状态变为失败，输出错误信息如下：

关键代码 #

self.audioSettings = @{
  AVFormatIDKey: @(kAudioFormatMPEG4AAC),
  AVNumberOfChannelsKey: @2,
  AVSampleRateKey: @44100,
  AVEncoderBitRateKey: @128000,
};

[self.audioInput requestMediaDataWhenReadyOnQueue:self.inputQueue usingBlock:^ {
  while (input.isReadyForMoreMediaData) {
    CMSampleBufferRef sampleBuffer = [output copyNextSampleBuffer];
    if (sampleBuffer) {
        if ([input appendSampleBuffer:sampleBuffer]) {
          // Something went wrong here！
        }
    }
  }
}];

错误信息 #

Error Domain=AVFoundationErrorDomain Code=-11800 "This operation cannot be completed" UserInfo={
  NSLocalizedFailureReason=An unknown error occurred（-12780）, 
  NSLocalizedDescription=This operation cannot be completed, 
  NSUnderlyingError=0x3023db8a0 {
    Error Domain=NSOSStatusErrorDomain Code=-12780 "(null)"
  }
}

成因分析 #

• iPhone 16 Pro Max录制的原生音频格式大概率是**kAudioFormatAppleLossless（ALAC）**或高规格PCM，你硬编码的音频参数（44.1kHz、双声道）和原始音频的实际参数不匹配，编码器无法兼容处理。
• 错误码-12780对应AVErrorInvalidSampleBuffer，说明传入的CMSampleBufferRef不符合AVAssetWriterInput的预期：要么是采样率、声道数、编码格式不匹配，要么是样本缓冲区的时间戳、配置信息存在异常。
• 代码未处理sampleBuffer为空的场景，且写入失败后没有及时终止写入流程并释放资源，导致写入器陷入异常状态。

修复方案 #

1. 匹配原始音频参数 #

不要硬编码音频参数，先获取原始视频音频轨道的实际参数，再对应设置编码参数：

// 获取原始音频轨道
AVAssetTrack *audioTrack = [[asset tracksWithMediaType:AVMediaTypeAudio] firstObject];
// 获取音频格式描述
NSDictionary *audioFormatDict = [audioTrack formatDescriptions].firstObject;
CMFormatDescriptionRef formatDesc = (__bridge CMFormatDescriptionRef)audioFormatDict;
const AudioStreamBasicDescription *asbd = CMAudioFormatDescriptionGetStreamBasicDescription(formatDesc);

// 基于原始参数生成编码设置
self.audioSettings = @{
    AVFormatIDKey: @(kAudioFormatMPEG4AAC),
    AVNumberOfChannelsKey: @(asbd->mChannelsPerFrame),
    AVSampleRateKey: @(asbd->mSampleRate),
    AVEncoderBitRateKey: @128000,
};

2. 处理音频格式转换 #

如果原始音频是ALAC等非PCM格式，需先转换为编码器支持的格式。可以在创建AVAssetReaderTrackOutput时指定输出格式：

AudioStreamBasicDescription outputASBD = {0};
outputASBD.mFormatID = kAudioFormatLinearPCM;
// 匹配原始采样率
outputASBD.mSampleRate = asbd->mSampleRate;
outputASBD.mChannelsPerFrame = asbd->mChannelsPerFrame;
outputASBD.mBitsPerChannel = 16;
outputASBD.mBytesPerPacket = outputASBD.mChannelsPerFrame * 2;
outputASBD.mBytesPerFrame = outputASBD.mChannelsPerFrame * 2;
outputASBD.mFramesPerPacket = 1;
outputASBD.mFormatFlags = kAudioFormatFlagIsSignedInteger | kAudioFormatFlagIsPacked;

CMFormatDescriptionRef outputFormatDesc = CMAudioFormatDescriptionCreate(kCFAllocatorDefault, &outputASBD, 0, NULL, NULL);
AVAssetReaderTrackOutput *audioOutput = [[AVAssetReaderTrackOutput alloc] initWithTrack:audioTrack outputSettings:(__bridge NSDictionary *)outputFormatDesc];
CFRelease(outputFormatDesc);

3. 完善错误处理逻辑 #

写入失败时及时获取错误信息，终止写入流程并释放资源：

if (sampleBuffer) {
    if (![input appendSampleBuffer:sampleBuffer]) {
        NSError *writeError = self.assetWriter.error;
        NSLog(@"音频写入失败：%@", writeError.localizedDescription);
        // 终止写入
        [self.assetWriter finishWritingWithCompletionHandler:^{
            // 处理后续清理逻辑
        }];
        break;
    }
    // 释放样本缓冲区
    CFRelease(sampleBuffer);
} else {
    // 无更多样本，结束输入
    [input markAsFinished];
    break;
}

4. 检查时间戳同步 #

确保音频样本的时间戳和视频样本在同一时间轴上，避免时间戳不连续或超出写入器时间范围：

if (sampleBuffer) {
    CMTime presentationTime = CMSampleBufferGetPresentationTimeStamp(sampleBuffer);
    // 可添加时间戳合法性检查
    if (CMTimeCompare(presentationTime, kCMTimeInvalid) == 0) {
        NSLog(@"无效的音频时间戳");
        CFRelease(sampleBuffer);
        continue;
    }
    // 执行写入操作
}

内容的提问来源于stack exchange，提问作者Mokyz