0

I'm having trouble with a AudioGraph in a UWP app. My trouble is that when I'm adding adding an Outgoing Connection to my input device node I get a Exception thrown with HRESULT: 0x88960001.

Using trial and error I figured out that the problem is with the MediaEncoding profile I set on both the input and output node... but I have now 2 settings only work on one machine and the other only works on another machine... I needed the audio to be Single Channel, 16bit sample with a 16K sample in PCM.

The code I'm using is fairly simple and was based on samples that exist online, the thing is it works only on some microphones and not others... I needed it to be generic and have always the same output so that I can input that to my service endpoint.

I left the 2 AudioEncodingProperties settings there (one of them commented out)

var result = await AudioGraph.CreateAsync(
          new AudioGraphSettings(AudioRenderCategory.Media));

        if (result.Status == AudioGraphCreationStatus.Success)
        {
            this.graph = result.Graph;

            var microphone = await DeviceInformation.CreateFromIdAsync(
              MediaDevice.GetDefaultAudioCaptureId(AudioDeviceRole.Default));

            // Low gives us 1 channel, 16-bits per sample, 16K sample rate.
            var outProfile = MediaEncodingProfile.CreateWav(AudioEncodingQuality.Low);
            //outProfile.Audio = AudioEncodingProperties.CreatePcm(16000, 1, 16);
            outProfile.Audio = AudioEncodingProperties.CreatePcm(44100, 2, 32);

            var inProfile = MediaEncodingProfile.CreateWav(AudioEncodingQuality.Low);
            //inProfile.Audio = AudioEncodingProperties.CreatePcm(16000, 1, 16);
            inProfile.Audio = AudioEncodingProperties.CreatePcm(44100, 2, 32);

            var outputResult = await this.graph.CreateFileOutputNodeAsync(file,
              outProfile);

            if (outputResult.Status == AudioFileNodeCreationStatus.Success)
            {
                this.outputNode = outputResult.FileOutputNode;

                var inputResult = await this.graph.CreateDeviceInputNodeAsync(
                  MediaCategory.Speech,
                  inProfile.Audio,
                  microphone);

                if (inputResult.Status == AudioDeviceNodeCreationStatus.Success)
                {
                    //EXEPTION IS THROWN HERE
                    inputResult.DeviceInputNode.AddOutgoingConnection(
                          this.outputNode);

                    this.graph.Start();
                }
            }
        }
        else
        {
            throw new Exception("Could not create AudioGraph");
        }

Thank you guys for our help

user3628320
  • 1
  • 1
    Probably 16Khz quality is not supported everywhere, but CD or DAT quality on the other hand will be. Solution would be to resample your audio before exporting it. – aybe Oct 15 '16 at 14:18

1 Answers1

0

Instead of constructing the inputResult with the inProfile.Audio object, try using the default encoding properties for the graph:

var enc = graph.EncodingProperties;
CreateAudioDeviceInputNodeResult deviceInputNodeResult = await graph.CreateDeviceInputNodeAsync(MediaCategory.Media, enc, this.outputNode);

I had a similar issue and this cleared things up, and allowed the audio to be input from any device without specifying the encoding profile.

memote
  • 1
  • 1