0

我正在编写一个模拟器，需要我不断地将生成的原始音频样本推送到QAudioOutput. 这是通过将字节写入底层来实现的，QIODevice如下所示：

Qt_speaker::Qt_speaker()
{
    QAudioFormat fmt;
    fmt.setSampleRate(44100);
    fmt.setChannelCount(1);
    fmt.setSampleSize(sizeof(uint8_t)*8);
    fmt.setCodec("audio/pcm");
    fmt.setByteOrder(QAudioFormat::LittleEndian);
    fmt.setSampleType(QAudioFormat::UnSignedInt);

    QAudioDeviceInfo info(QAudioDeviceInfo::defaultOutputDevice());
    if (!info.isFormatSupported(fmt)) {
        qWarning() << "Raw audio format not supported by backend, cannot play audio.\n"
                      "If you're running a Debian system, try manually installing Qt's"
                      "multimedia plugins with\n"
                      "\t'sudo apt-get install libqt5multimedia5-plugins'";
        toggle(false);
        return;
    }
    output_ = new QAudioOutput(fmt, nullptr);
    output_->setBufferSize(44100);
    //connect(output_, SIGNAL(stateChanged(QAudio::State)),
            //this, SLOT(output_state_changed(QAudio::State)));
    device_ = output_->start();

}

void Qt_speaker::push_samples(const gameboy::Raw_audio &a)
{
    device_->write(reinterpret_cast<const char *>(a.data()), a.size());
}

我想跟踪排队的样本数量，因为声音会驱动仿真速度，从而驱动生成的原始音频样本。尝试调用QIODevice::read()总是返回 0。我该怎么做？

4

1 回答 1

-1

我认为您应该通知 QAudioInput 的信号，以便根据 QAudioInput “消耗”的样本行事。

于 2020-12-18T06:35:28.930 回答