"Read Failed" when capturing audio on linux (ReSpeaker Core V2) about psi HOT 2 OPEN

It looks like perhaps the device doesn't support single-channel. The AudioCapture component takes an optional AudioCaptureConfiguration with a WaveFormat which defaults to 16KHz, 1 channel, 16-bit PCM. You could try passing a config with different settings.

I'm not sure how to query the hardware for this, but one possible way would be to record a clip with arecord and then look at file details with mediainfo which does show the sampling rate, channel count, etc.

from psi.

I think you are pointing in the right direction...
respeaker@E3BC1:~$ arecord -Dhw:0,0 -f S16_LE -r 16000 -c 1 hello.wav
Recording WAVE 'hello.wav' : Signed 16 bit Little Endian, Rate 16000 Hz, Mono
arecord: set_params:1305: Channels count non available

When I use stereo, it works fine...
respeaker@E3BC1:~$ arecord -Dhw:0,0 -f S16_LE -r 16000 -c 2 hello.wav
Recording WAVE 'hello.wav' : Signed 16 bit Little Endian, Rate 16000 Hz, Stereo
^CAborted by signal Interrupt...

But using two channels produces another error... but interestingly not an exception. it just prints "Seqmentation faul" to the console and stops...
respeaker@E3BC1:/mnt/Test$ ./ConsoleApp3
Please enter the device name
plughw:0,0
Please enter the sampling rate
16000
Please enter the channel count
2
Segmentation fault
respeaker@E3BC1:/mnt/Test$

Here is the code from my test app (xxx is a SimpleConsumer that does nothing else than printing a message to the console in "Receive"):
using (var pipeline = Microsoft.Psi.Pipeline.Create())
{
//var devices = AudioCapture.GetAvailableDevices();

            //foreach (var device in devices)
            //{
            //    Console.WriteLine(device);
            //}

            //Console.ReadLine();

            //audioCaptureConfiguration.OptimizeForSpeech = true;
            //audioCaptureConfiguration.OutputFormat = Microsoft.Psi.Audio.WaveFormat.Create16kHz1Channel16BitPcm();

            var audioCaptureConfiguration = new Microsoft.Psi.Audio.AudioCaptureConfiguration();

            Console.WriteLine("Please enter the device name");
            audioCaptureConfiguration.DeviceName = Console.ReadLine();

            Console.WriteLine("Please enter the sampling rate");
            var samplingRate = int.Parse(Console.ReadLine());

            Console.WriteLine("Please enter the channel count");
            var channelCount = int.Parse(Console.ReadLine());

            audioCaptureConfiguration.Format = Microsoft.Psi.Audio.WaveFormat.CreatePcm(samplingRate, 16, channelCount);

            using (var audioCapture = new Microsoft.Psi.Audio.AudioCapture(pipeline, audioCaptureConfiguration))
            {
                var audioInput = (Microsoft.Psi.IProducer<Microsoft.Psi.Audio.AudioBuffer>)audioCapture;

                audioInput.PipeTo(new xxx(pipeline));

                //pipeline.Run();

                using (pipeline.RunAsync())
                {
                    Console.ReadLine();
                }
            }
        }

What does Seqmentation fault mean?

Perhaps this could help...
https://stackoverflow.com/questions/48588513/segmentation-fault-in-alsa-capturing-example

from psi.