I have a WAV file that I decode with the QAudioDecoder. As a result I have a QAudioBuffer object. I want to store the data stored in QAudioBuffer in a QByteArray for my QIODevice derived class. I want to use this data in the ReadData method of my derived class for audio output. I now have 2 questions:
Thanks in advance.
Here is the code
QAudioFormat *format_decoder;
format_decoder = new QAudioFormat;
format_decoder->setSampleRate(44100);
format_decoder->setChannelCount(1);
format_decoder->setSampleFormat(QAudioFormat::Int16);
QAudioDecoder decoder;
decoder.setSource(filenameSource);
decoder.setAudioFormat(*format_decoder);
decoder.start();
QObject::connect(&decoder, &QAudioDecoder::bufferReady, this, &MainWindow::slot_bufReady);
and the slot
void MainWindow::slot_bufReady(){
QAudioBuffer buffer = m_audioDecoder->read();
QByteArray buffer_ByteArray(buffer.constData<char>(), buffer.byteCount());
QFile file(filenameTest1);
if(!file.open(QIODevice::WriteOnly|QIODevice::Append)) {
qDebug() << "ERRO "; }
QTextStream strem(&file);
for(auto const dat: buffer_ByteArray) {
strem<< qreal(dat)/128.0<< "\r\n";
}
file.cloe();
This looks suspicious:
for(auto const dat: buffer_ByteArray) {
strem<< qreal(dat)/128.0<< "\r\n";
}
Your audio format is 16-bit mono. Reading it byte by byte is a non-starter. Read it sample by sample. That is, read two bytes at a time and convert. More likely this:
int16_t* data = (int16_t*)(buffer.data());
int samples = buffer.sampleCount();
for (int i = 0; i < samples; i++)
{
strem << data[i] << "\r\n";
}
The above will save your samples into a text file. You could plot it with Excel. But as others have said, that's not as useful as saving in as binary. You could prepend a WAV file header such that it can be played and analyzed with other tools.
Update
If your intent is to transcode from 16-bit to 8-bit, this is how you would likely do it:
int16_t* data = (int16_t*)(buffer.data());
QByteArray buffer_ByteArray(buffer.sampleCount(), '\0');
for (size_t i = 0; i < samples; i++) {
buffer_ByteArray[i] = (char)(data[i] / 256); // 16-bit to 8-bit
}
Note: some audio platforms use unsigned integers for 8-bit audio. That is the zero amplitude sample is 0x80. This is the case for 8-bit WAV files. If that's in play, then change this line:
buffer_ByteArray[i] = (char)(data[i] / 256); // 16-bit to 8-bit
To this:
char c = (char)(data[i] / 256); // 16-bit to 8-bit signed
const unsigned char mask = 0x80;
buffer_ByteArray[i] = (char)(mask ^ c);