Android Wear Audio Recorder using the ChannelAPI

Question

I'm trying to build an audio recorder app for Android Wear. Right now, I'm able to capture the audio on the watch, stream it to phone and save it on a file. However, the audio file is presenting gaps or cropped parts.

I found this aswered questions related to my problem link1, link2, but they couldn't help me.

---

Here is my code:

First, on the watch side, I create the channel using the channelAPI and sucessfully send the audio being captured on the watch to the smartphone.

//here are the variables values that I used

//44100Hz is currently the only rate that is guaranteed to work on all devices
//but other rates such as 22050, 16000, and 11025 may work on some devices.

private static final int RECORDER_SAMPLE_RATE = 44100; 
private static final int RECORDER_CHANNELS = AudioFormat.CHANNEL_IN_MONO;
private static final int RECORDER_AUDIO_ENCODING = AudioFormat.ENCODING_PCM_16BIT;
int BufferElements2Rec = 1024; 
int BytesPerElement = 2; 

//start the process of recording audio
private void startRecording() {

    recorder = new AudioRecord(MediaRecorder.AudioSource.MIC,
            RECORDER_SAMPLE_RATE, RECORDER_CHANNELS,
            RECORDER_AUDIO_ENCODING, BufferElements2Rec * BytesPerElement);

    recorder.startRecording();
    isRecording = true;
    recordingThread = new Thread(new Runnable() {
        public void run() {
            writeAudioDataToPhone();
        }
    }, "AudioRecorder Thread");
    recordingThread.start();
}

private void writeAudioDataToPhone(){

    short sData[] = new short[BufferElements2Rec];
    ChannelApi.OpenChannelResult result = Wearable.ChannelApi.openChannel(googleClient, nodeId, "/mypath").await();
    channel = result.getChannel();

    Channel.GetOutputStreamResult getOutputStreamResult = channel.getOutputStream(googleClient).await();
    OutputStream outputStream = getOutputStreamResult.getOutputStream();

    while (isRecording) {
        // gets the voice output from microphone to byte format

        recorder.read(sData, 0, BufferElements2Rec);
        try {
            byte bData[] = short2byte(sData);
            outputStream.write(bData);
        } catch (IOException e) {
            e.printStackTrace();
        }
    }
    try {
        outputStream.close();
    } catch (IOException e) {
        e.printStackTrace();
    }
}

Then, on the smartphone side, I receive the audio data from the channel and write it to a PCM file.

public void onChannelOpened(Channel channel) {
    if (channel.getPath().equals("/mypath")) {
        Channel.GetInputStreamResult getInputStreamResult = channel.getInputStream(mGoogleApiClient).await();
        inputStream = getInputStreamResult.getInputStream();

        writePCMToFile(inputStream);

        MainActivity.this.runOnUiThread(new Runnable() {
            public void run() {
                Toast.makeText(MainActivity.this, "Audio file received!", Toast.LENGTH_SHORT).show();
            }
        });
    }
}

public void writePCMToFile(InputStream inputStream) {
    OutputStream outputStream = null;

    try {
        // write the inputStream to a FileOutputStream
        outputStream = new FileOutputStream(new File("/sdcard/wearRecord.pcm"));

        int read = 0;
        byte[] bytes = new byte[1024];

        while ((read = inputStream.read(bytes)) != -1) {
            outputStream.write(bytes, 0, read);
        }

        System.out.println("Done writing PCM to file!");

    } catch (Exception e) {
        e.printStackTrace();
    } finally {
        if (inputStream != null) {
            try {
                inputStream.close();
            } catch (Exception e) {
                e.printStackTrace();
            }
        }
        if (outputStream != null) {
            try {
                // outputStream.flush();
                outputStream.close();
            } catch (Exception e) {
                e.printStackTrace();
            }

        }
    }
}

---

What am I doing wrong or what are your suggestions to achieve a perfect gapless audio file on the smartphone? Thanks in advance.

Answer 1

I noticed in your code that you are reading everything into a short[] array, and then converting it to a byte[] array for the Channel API to send. Your code also creates a new byte[] array through each iteration of the loop, which will create a lot of work for the garbage collector. In general, you want to avoid allocations inside loops.

I would allocate one byte[] array at the top, and let the AudioRecord class store it directly into the byte[] array (just make sure you allocate twice as many bytes as you did shorts), with code like this:

mAudioTemp = new byte[bufferSize];

int result;
while ((result = mAudioRecord.read(mAudioTemp, 0, mAudioTemp.length)) > 0) {
  try {
    mAudioStream.write(mAudioTemp, 0, result);
  } catch (IOException e) {
    Log.e(Const.TAG, "Write to audio channel failed: " + e);
  }
}

I also tested this with a 1 second audio buffer, using code like this, and it worked nicely. I'm not sure what the minimum buffer size is before it starts to have problems:

int bufferSize = Math.max(
  AudioTrack.getMinBufferSize(44100, AudioFormat.CHANNEL_OUT_MONO, AudioFormat.ENCODING_PCM_16BIT),
  44100 * 2);