Multichannel audio output with QAudioOutput Class

[Franky01]

Hi all,

I am a student of electronics and mechatronics engineering and currently working on an audio project. I build a 8-channel linear speaker array as a demonstrator for speaker beamforming. I have a multichannel soundcard to handle the signal output and the input for the amplifiers.

It's the first time I'm using Qt (on a windows system with Qt Creator), but I managed to write a working testcode. This small app outputs a simple sinewave via the QAudioOutput Class.

As the next step, I want to output sinewaves on multiple channels at the same time.
I red the documentation of the QAudioOutput Class, but there is just a section for setting the channel count. I didn't find any info on how to pass the multichannel buffer to the output device.

Is it possible to use a multidimensional data array to store the data points of the sinewave in (e.g. first row of data is channel 1, second row is channel 2 and so on)?
Or is it possible to concatenate multiple QByteArrays to one multidimensional buffer?

Thank you.
Kind regards, Markus.

[SGaist]

Hi and welcome to devnet,

What about writing to the QIODevice returned by the start method ?

[Franky01]

Hi, I am back at the project again.

What about writing to the QIODevice returned by the start method ?

Yes thats how I do it. See reduced code below.

//set audio output format unsing QAudioFormat
//...
//QBuffer with sine data (stored in ByteArray)
QBuffer* input = new QBuffer(byteBuffer);
input->open(QIODevice::ReadOnly);
// Create an audio output with our QAudioFormat
QAudioOutput* audio = new QAudioOutput(audioFormat, this);
//play audio
audio->start(input);

The question is: if I use more than one channel, in which format do I have to send the data respectively how do I have to store multichannel data in a ByteArray?
Is it possible to use a ByteArray for each audio channel or use a multidimensional array for the channels?
Or do I have to concatenate the multichannel data like sample1_ch1 - sample1_ch2 - sample1_ch3 ....sample_n-ch_m in a single dimensional array?

Thanks for your help.
Kind regards.

[SGaist]

You should interleave the audio data as required by the number of channels.