[修正]IOS中使用SoundTouch库实现变声

之前写过一片文章“IOS中使用SoundTouch库实现变声”。

最后有很多朋友在博客上说发出来的代码在模拟器上运行可以，不能在真机上运行，前段时间一直也很忙，2013好看的美剧排行榜没时间翻代码出来查看，望谅解。

今天抽空检查了一下问题的根源，并且在IPhone 4（ios5.0.1）上测试通过，下面更新一些代码。

 

同样是Apple官方的SpeakHere，要下载代码的，请查看上一篇文章。

void AQRecorder::StartRecord(CFStringRef inRecordFile)
{
mSoundTouch.setSampleRate(44100);//mRecordFormat.mSampleRate
    mSoundTouch.setChannels(1);//mRecordFormat.mChannelsPerFrame
    mSoundTouch.setTempoChange(1.0);
    mSoundTouch.setPitchSemiTones(9);
    mSoundTouch.setRateChange(-0.7);

    mSoundTouch.setSetting(SETTING_SEQUENCE_MS, 40);
    mSoundTouch.setSetting(SETTING_SEEKWINDOW_MS, 16);
    mSoundTouch.setSetting(SETTING_OVERLAP_MS, 8);

    //Only use one of the following two options
    //  mSoundTouch.setSetting(SETTING_USE_QUICKSEEK, 0);
    //  mSoundTouch.setSetting(SETTING_USE_AA_FILTER, !(0));
    //  mSoundTouch.setSetting(SETTING_AA_FILTER_LENGTH, 32);

 

红色的两行很重要，需要指明采样率和声道。如果需要使用后面注释两个变量，需要在SetupAudioFormat方法执行之后才可以，否则无效，原因大家都懂的。

将麦克风捕捉的声音回调函数按照一下代码更新。

// ____________________________________________________________________________________
// AudioQueue callback function, called when an input buffers has been filled.
void AQRecorder::MyInputBufferHandler(  void *                              inUserData,
                                        AudioQueueRef                       inAQ,
                                        AudioQueueBufferRef                 inBuffer,
                                        const AudioTimeStamp *              inStartTime,
                                        UInt32                              inNumPackets,
                                        const AudioStreamPacketDescription* inPacketDesc)
{

    AQRecorder *aqr = (AQRecorder *)inUserData;
    try {
        if (inNumPackets > 0) {
            UInt32 audioDataByteSize = inBuffer->mAudioDataByteSize;
            CAStreamBasicDescription queueFormat = aqr->DataFormat();
            SoundTouch *soundTouch = aqr->GetSoundTouch();

            uint nSamples = audioDataByteSize/queueFormat.mBytesPerPacket;
            soundTouch->putSamples((const SAMPLETYPE *)inBuffer->mAudioData,nSamples);

            SAMPLETYPE *samples = (SAMPLETYPE *)malloc(audioDataByteSize);
            UInt32 numSamples;
            do {
                memset(samples, 0, audioDataByteSize);
                numSamples = soundTouch->receiveSamples((SAMPLETYPE *)samples, nSamples);
                // write packets to file
                XThrowIfError(AudioFileWritePackets(aqr->mRecordFile,
                                                    FALSE,
                                                    numSamples*queueFormat.mBytesPerPacket,
                                                    NULL,
                                                    aqr->mRecordPacket,
                                                    &numSamples,
                                                    samples),
                              "AudioFileWritePackets failed");
                aqr->mRecordPacket += numSamples;
            } while (numSamples!=0);
            free(samples);
        }

        // if we're not stopping, re-enqueue the buffe so that it gets filled again
        if (aqr->IsRunning())
            XThrowIfError(AudioQueueEnqueueBuffer(inAQ, inBuffer, 0, NULL), "AudioQueueEnqueueBuffer failed");
    } catch (CAXException e) {
        char buf[256];
        fprintf(stderr, "Error: %s (%s)\n", e.mOperation, e.FormatError(buf));
    }
}

原因大家自己应该能看明白。

之前是在xcode 3.x上写的代码，刚开始有朋友反馈时，我以为是版本问题产生的，结果不是神奇影院.

本次更新的代码是我在xcode4.2和iphone4上测试无误的，如果还有问题的，可以仔细看看前后两篇文章。不要吧参数设置错误了。

ps：QQ经常不在线，有问题上微博：http://weibo.com/yarin

本文出自 “yarin's blog™” 博客，请务必保留此出处http://yarin.blog.51cto.com/1130898/794768