Ios Avassetrader获取的音频样本数量不正确

我试图从视频中获取音频曲目作为原始pcm数据。计划是将此数据传递到libpd中的float*表数组中。我已设法获得一些样本数据，但报告的样本数量太少。例如，对于一个29秒的剪辑，我得到了报告的3968个样本。我期待的是每个样本的振幅。对于29秒音频长度的音轨，我希望阵列大小为1278900（44.1khz）

以下是我根据其他示例总结的内容：

- (void)audioCapture {

NSLog(@"capturing");
NSError *error;

AVAssetReader* reader = [[AVAssetReader alloc] initWithAsset:self.videoAsset error:&error];

NSArray *audioTracks = [self.videoAsset tracksWithMediaType:AVMediaTypeAudio];

AVAssetTrack *audioTrack = nil;

if ([audioTracks count] > 0)
    audioTrack = [audioTracks objectAtIndex:0];

// Decompress to Linear PCM with the asset reader
NSDictionary *decompressionAudioSettings = [NSDictionary dictionaryWithObjectsAndKeys:
                                            [NSNumber numberWithUnsignedInt:kAudioFormatLinearPCM], AVFormatIDKey,
                                            nil];

AVAssetReaderOutput *output = [AVAssetReaderTrackOutput assetReaderTrackOutputWithTrack:audioTrack outputSettings:decompressionAudioSettings];
[reader  addOutput:output];


[reader startReading];

CMSampleBufferRef sample = [output copyNextSampleBuffer];

//array for our sample amp values
SInt16* samples = NULL;

//sample count;
CMItemCount numSamplesInBuffer = 0;

//sample buffer
CMBlockBufferRef buffer;

while( sample != NULL )
{
    sample = [output copyNextSampleBuffer];

    if( sample == NULL )
        continue;

    buffer = CMSampleBufferGetDataBuffer( sample );

    size_t lengthAtOffset;
    size_t totalLength;
    char* data;

    if( CMBlockBufferGetDataPointer( buffer, 0, &lengthAtOffset, &totalLength, &data ) != noErr )
    {
        NSLog( @"error!" );
        break;
    }

    numSamplesInBuffer = CMSampleBufferGetNumSamples(sample);

    AudioBufferList audioBufferList;

    CMSampleBufferGetAudioBufferListWithRetainedBlockBuffer(
                                                            sample,
                                                            NULL,
                                                            &audioBufferList,
                                                            sizeof(audioBufferList),
                                                            NULL,
                                                            NULL,
                                                            kCMSampleBufferFlag_AudioBufferList_Assure16ByteAlignment,
                                                            &buffer
                                                            );

    for (int bufferCount=0; bufferCount < audioBufferList.mNumberBuffers; bufferCount++) {
        samples = (SInt16 *)audioBufferList.mBuffers[bufferCount].mData;


        NSLog(@"idx %f",(double)samples[bufferCount]);
    }
}
NSLog(@"num samps in buf %ld",numSamplesInBuffer);

//[PdBase copyArray:(float *)samples 
     //toArrayNamed:@"DestTable" withOffset:0    
            //count:pdTableSize];

-（无效）音频捕获{
NSLog（@“捕获”）；
n错误*错误；
Avassetrader*reader=[[Avassetrader alloc]initWithAsset:self.videoAsset错误：&error]；
NSArray*audioTracks=[self.videoAsset tracksWithMediaType:AVMediaTypeAudio]；
AVAssetTrack*音轨=零；
如果（[audioTracks count]>0）
audioTrack=[audioTracks对象索引：0]；
//使用资产读取器解压缩到线性PCM
NSDictionary*解压缩音频设置=[NSDictionary Dictionary WithObjectsAndKeys:
[NSNumber numberWithUnsignedInt:kAudioFormatLinearPCM]，AVFormatIDKey，
零]；
AVAssetReaderOutput*输出=[AVAssetReaderTrackOutput assetReaderTrackOutputWithTrack:audioTrack输出设置：解压缩音频设置]；
[读卡器添加输出：输出]；
[读者开始阅读]；
CMSampleBufferRef样本=[output copyNextSampleBuffer]；
//我们的样本安培值数组
SInt16*样本=NULL；
//样本计数；
CMItemCount numSamplesInBuffer=0；
//样品缓冲液
CMBlockBufferRef缓冲区；
while（示例！=NULL）
{
样本=[output copyNextSampleBuffer]；
if（sample==NULL）
继续；
缓冲区=CMSampleBufferGetDataBuffer（样本）；
尺寸/长度偏移量；
尺寸/总长度；
字符*数据；
if（CMBlockBufferGetDataPointer（缓冲区、0和长度偏移量、总长度和数据）！=noErr）
{
NSLog（@“错误！”）；
打破
}
numSamplesInBuffer=CMSampleBufferGetNumSamples（样本）；
AudioBufferList AudioBufferList；
CMSampleBufferGetAudioBufferListWithRetainedBlockBuffer(
样品
无效的
&音频缓冲列表，
sizeof（音频缓冲列表），
无效的
无效的
kCMSampleBufferFlag_Audio BufferList_确保16通过对齐，
&缓冲区
);
对于（int bufferCount=0；bufferCount

}