【Android多媒体（重采样与混频）】重采样等基本概念

重采样和混频的概念

关于音频采样，百度上面找到这样的解答：

频率对应于时间轴线，振幅对应于电平轴线。波是无限光滑的，弦线可以看成由无数点组成，由于存储空间是相对有限的，数字编码过程中，必须对弦线的点进行采样。采样的过程就是抽取某点的频率值，很显然，在一秒中内抽取的点越多，获取得频率信息更丰富，为了复原波形，一次振动中，必须有2个点的采样，人耳能够感觉到的最高频率为20kHz，因此要满足人耳的听觉要求，则需要至少每秒进行40k次采样，用40kHz表达，这个40kHz就是采样率。我们常见的CD，采样率为44.1kHz。光有频率信息是不够的，我们还必须获得该频率的能量值并量化，用于表示信号强度。量化电平数为2的整数次幂，我们常见的CD位16bit的采样大小，即2的16次方。采样大小相对采样率更难理解，因为要显得抽象点，举个简单例子：假设对一个波进行8次采样，采样点分别对应的能量值分别为A1-A8，但我们只使用2bit的采样大小，结果我们只能保留A1-A8中4个点的值而舍弃另外4个。如果我们进行3bit的采样大小，则刚好记录下8个点的所有信息。采样率和采样大小的值越大，记录的波形更接近原始信号

而重采样，就是为了满足播放与存储的需求，使用某种插值或者抽取的算法，改变采样率的过程。

Android系统中间的重采样与混频

再android多媒体系统中间：
录音线程RecordThread直接调用AudioResampler进行重采样
播放线程分为两种情况：

• 硬解码播放线程OffloadThread不会对音频进行重采样
• 非硬解码使用MixerThread线程播放，将会对音频按照需要使用AudioMixer进行重采样，实际上AudioMixer也是调用AudioResampler进行重采样工作。
  -

从AudioMixer提供的接口方法分析

主要作用是重采样，将各个音轨的数据混合在一起输出给音频设备

主要通过接收参数，执行相关操作，
参考路径：
frameworks/av/services/audioflinger/AudioMixer.h
参数设置接口：
setParameter(int name, int target, int param, void *value);

  enum { // names        // track names (MAX_NUM_TRACKS units)        TRACK0          = 0x1000,        // 0x2000 is unused        // setParameter targets        TRACK           = 0x3000,        RESAMPLE        = 0x3001,        RAMP_VOLUME     = 0x3002, // ramp to new volume        VOLUME          = 0x3003, // don't ramp        TIMESTRETCH     = 0x3004,        // set Parameter names        // for target TRACK        CHANNEL_MASK    = 0x4000,        FORMAT          = 0x4001,        MAIN_BUFFER     = 0x4002,        AUX_BUFFER      = 0x4003,        DOWNMIX_TYPE    = 0X4004,        MIXER_FORMAT    = 0x4005, // AUDIO_FORMAT_PCM_(FLOAT|16_BIT)        MIXER_CHANNEL_MASK = 0x4006, // Channel mask for mixer output        // for target RESAMPLE        SAMPLE_RATE     = 0x4100, // Configure sample rate conversion on this track name;                                  // parameter 'value' is the new sample rate in Hz.                                  // Only creates a sample rate converter the first time that                                  // the track sample rate is different from the mix sample rate.                                  // If the new sample rate is the same as the mix sample rate,                                  // and a sample rate converter already exists,                                  // then the sample rate converter remains present but is a no-op.        RESET           = 0x4101, // Reset sample rate converter without changing sample rate.                                  // This clears out the resampler's input buffer.        REMOVE          = 0x4102, // Remove the sample rate converter on this track name;                                  // the track is restored to the mix sample rate.        // for target RAMP_VOLUME and VOLUME (8 channels max)        // FIXME use float for these 3 to improve the dynamic range        VOLUME0         = 0x4200,        VOLUME1         = 0x4201,        AUXLEVEL        = 0x4210,        // for target TIMESTRETCH        PLAYBACK_RATE   = 0x4300, // Configure timestretch on this track name;                                  // parameter 'value' is a pointer to the new playback rate.    };

从AudioResampler提供的接口方法分析

从构造函数来看，就是根据不同的音质要求，创建不同的AudioResampler子类实例。如下枚举可见，重采样方法有三种插值方法

   // Determines quality of SRC.    //  LOW_QUALITY: linear interpolator (1st order)    //  MED_QUALITY: cubic interpolator (3rd order)    //  HIGH_QUALITY: fixed multi-tap FIR (e.g. 48KHz->44.1KHz)    // NOTE: high quality SRC will only be supported for    // certain fixed rate conversions. Sample rate cannot be    // changed dynamically.    enum src_quality {        DEFAULT_QUALITY=0,        LOW_QUALITY=1,        MED_QUALITY=2,        HIGH_QUALITY=3,        VERY_HIGH_QUALITY=4,        DYN_LOW_QUALITY=5,        DYN_MED_QUALITY=6,        DYN_HIGH_QUALITY=7,#ifdef QTI_RESAMPLER        QTI_QUALITY=8,#endif    };

构造函数

AudioResampler* AudioResampler::create(audio_format_t format, int inChannelCount,        int32_t sampleRate, src_quality quality) {    bool atFinalQuality;    if (quality == DEFAULT_QUALITY) {        // read the resampler default quality property the first time it is needed        int ok = pthread_once(&once_control, init_routine);        if (ok != 0) {            ALOGE("%s pthread_once failed: %d", __func__, ok);        }        quality = defaultQuality;        atFinalQuality = false;    } else {        atFinalQuality = true;    }    /* if the caller requests DEFAULT_QUALITY and af.resampler.property     * has not been set, the target resampler quality is set to DYN_MED_QUALITY,     * and allowed to "throttle" down to DYN_LOW_QUALITY if necessary     * due to estimated CPU load of having too many active resamplers     * (the code below the if).     */    if (quality == DEFAULT_QUALITY) {        quality = DYN_MED_QUALITY;    }    // naive implementation of CPU load throttling doesn't account for whether resampler is active    pthread_mutex_lock(&mutex);    for (;;) {        uint32_t deltaMHz = qualityMHz(quality);        uint32_t newMHz = currentMHz + deltaMHz;        if ((qualityIsSupported(quality) && newMHz <= maxMHz) || atFinalQuality) {            ALOGV("resampler load %u -> %u MHz due to delta +%u MHz from quality %d",                    currentMHz, newMHz, deltaMHz, quality);            currentMHz = newMHz;            break;        }        // not enough CPU available for proposed quality level, so try next lowest level        switch (quality) {        default:        case LOW_QUALITY:            atFinalQuality = true;            break;        case MED_QUALITY:            quality = LOW_QUALITY;            break;        case HIGH_QUALITY:            quality = MED_QUALITY;            break;        case VERY_HIGH_QUALITY:            quality = HIGH_QUALITY;            break;        case DYN_LOW_QUALITY:            atFinalQuality = true;            break;        case DYN_MED_QUALITY:            quality = DYN_LOW_QUALITY;            break;        case DYN_HIGH_QUALITY:            quality = DYN_MED_QUALITY;            break;#ifdef QTI_RESAMPLER        case QTI_QUALITY:            quality = DYN_HIGH_QUALITY;            break;#endif        }    }    pthread_mutex_unlock(&mutex);    AudioResampler* resampler;    switch (quality) {    default:    case LOW_QUALITY:        ALOGV("Create linear Resampler");        LOG_ALWAYS_FATAL_IF(format != AUDIO_FORMAT_PCM_16_BIT);        resampler = new AudioResamplerOrder1(inChannelCount, sampleRate);        break;    case MED_QUALITY:        ALOGV("Create cubic Resampler");        LOG_ALWAYS_FATAL_IF(format != AUDIO_FORMAT_PCM_16_BIT);        resampler = new AudioResamplerCubic(inChannelCount, sampleRate);        break;    case HIGH_QUALITY:        ALOGV("Create HIGH_QUALITY sinc Resampler");        LOG_ALWAYS_FATAL_IF(format != AUDIO_FORMAT_PCM_16_BIT);        resampler = new AudioResamplerSinc(inChannelCount, sampleRate);        break;    case VERY_HIGH_QUALITY:        ALOGV("Create VERY_HIGH_QUALITY sinc Resampler = %d", quality);        LOG_ALWAYS_FATAL_IF(format != AUDIO_FORMAT_PCM_16_BIT);        resampler = new AudioResamplerSinc(inChannelCount, sampleRate, quality);        break;    case DYN_LOW_QUALITY:    case DYN_MED_QUALITY:    case DYN_HIGH_QUALITY:        ALOGV("Create dynamic Resampler = %d", quality);        if (format == AUDIO_FORMAT_PCM_FLOAT) {            resampler = new AudioResamplerDyn<float, float, float>(inChannelCount,                    sampleRate, quality);        } else {            LOG_ALWAYS_FATAL_IF(format != AUDIO_FORMAT_PCM_16_BIT);            if (quality == DYN_HIGH_QUALITY) {                resampler = new AudioResamplerDyn<int32_t, int16_t, int32_t>(inChannelCount,                        sampleRate, quality);            } else {                resampler = new AudioResamplerDyn<int16_t, int16_t, int32_t>(inChannelCount,                        sampleRate, quality);            }        }        break;#ifdef QTI_RESAMPLER    case QTI_QUALITY:        ALOGV("Create QTI_QUALITY Resampler = %d",quality);        resampler = new AudioResamplerQTI(format, inChannelCount, sampleRate);        break;#endif    }    // initialize resampler    resampler->init();    return resampler;}

之后将会从具体的播放与录音流程进行分析

多谢各位纠错！~
参考文章
http://blog.csdn.net/augusdi/article/details/12516337
http://blog.csdn.net/u010681466/article/details/50573440