PCM Wave file - stereo to mono
Asked Answered
X

3

9

I have an audio file which is stereo. Is converting it to mono just a case of skipping every other byte (after the header)? It's encoded in 16bit signed PCM format. I've got javax.sound.sampled available.

Here's code I tried that didn't work:

WaveFileWriter wfw = new WaveFileWriter();
AudioFormat format = new AudioFormat(Encoding.PCM_SIGNED, 44100, 16, 2, 2, 44100, false);
AudioFormat monoFormat = new AudioFormat(Encoding.PCM_SIGNED, 44100, 16, 1, 2, 44100, false);

byte[] audioData = dataout.toByteArray();
int length = audioData.length;
ByteArrayInputStream bais = new ByteArrayInputStream(audioData);

AudioInputStream stereoStream = new AudioInputStream(bais,format,length);
AudioInputStream monoStream = new AudioInputStream(stereoStream,format,length/2);

wfw.write(monoStream, Type.WAVE, new File(Environment.
                 getExternalStorageDirectory().getAbsolutePath()+"/stegDroid/un-ogged.wav"));

This code is used after reading a .ogg file using Jorbis to convert it to PCM data. The only problem is the result is stereo and I need it to be mono, so if there's another solution I'm happy to hear it!

Xl answered 17/1, 2011 at 15:1 Comment(0)
S
10

I have an audio file which is stereo. Is converting it to mono just a case of skipping every other byte (after the header)?

Almost - you want to skip every other sample, not byte. In your case it looks like each sample is of size 16 bits = 2 bytes. So you would want to take 2 bytes, skip 2 bytes, take 2 bytes and so on.

AudioInputStream monoStream = new AudioInputStream(stereoStream,format,length/2);

wfw.write(monoStream, Type.WAVE, new File(Environment.getExternalStorageDirectory().getAbsolutePath()+"/stegDroid/un-ogged.wav"));

This looks like you just write out the first half of the file instead of writing out every other sample. Also you have to fix the WAV header, to specify a single channel (see your monoFormat).

Swiercz answered 17/1, 2011 at 16:36 Comment(1)
If you skip every other sample then you're only getting ONE of the stereo (left/right) channels. You probably want to combine the left/right into a single mono channel. A naive way would be simply adding the samples together and dividing by 2 (averaging). Be careful to use appropriate types to avoid overflow. In the case of 16bit samples, make sure to use a 32bit int for the averaging. Worth noting that there are other ways to mix samples.Reparable
C
4

Take a look at this code. It helped me when I needed to mess with the bytes in a wav file.


package GlobalUtilities;

import java.applet.Applet;
import java.applet.AudioClip;
import java.net.URISyntaxException;
import java.util.logging.Level;
import java.util.logging.Logger;
import java.io.*;
import java.io.File;
import java.net.MalformedURLException;
import java.net.URL;
import javax.sound.sampled.*;

/**
 * This class handles the reading, writing, and playing of wav files. It is
 * also capable of converting the file to its raw byte [] form.
 *
 * based on code by Evan Merz modified by Dan Vargo
 * @author dvargo
 */
public class Wav {
    /*
    WAV File Specification
    FROM http://ccrma.stanford.edu/courses/422/projects/WaveFormat/
    The canonical WAVE format starts with the RIFF header:
    0         4   ChunkID          Contains the letters "RIFF" in ASCII form
    (0x52494646 big-endian form).
    4         4   ChunkSize        36 + SubChunk2Size, or more precisely:
    4 + (8 + SubChunk1Size) + (8 + SubChunk2Size)
    This is the size of the rest of the chunk
    following this number.  This is the size of the
    entire file in bytes minus 8 bytes for the
    two fields not included in this count:
    ChunkID and ChunkSize.
    8         4   Format           Contains the letters "WAVE"
    (0x57415645 big-endian form).

    The "WAVE" format consists of two subchunks: "fmt " and "data":
    The "fmt " subchunk describes the sound data's format:
    12        4   Subchunk1ID      Contains the letters "fmt "
    (0x666d7420 big-endian form).
    16        4   Subchunk1Size    16 for PCM.  This is the size of the
    rest of the Subchunk which follows this number.
    20        2   AudioFormat      PCM = 1 (i.e. Linear quantization)
    Values other than 1 indicate some
    form of compression.
    22        2   NumChannels      Mono = 1, Stereo = 2, etc.
    24        4   SampleRate       8000, 44100, etc.
    28        4   ByteRate         == SampleRate * NumChannels * BitsPerSample/8
    32        2   BlockAlign       == NumChannels * BitsPerSample/8
    The number of bytes for one sample including
    all channels. I wonder what happens when
    this number isn't an integer?
    34        2   BitsPerSample    8 bits = 8, 16 bits = 16, etc.

    The "data" subchunk contains the size of the data and the actual sound:
    36        4   Subchunk2ID      Contains the letters "data"
    (0x64617461 big-endian form).
    40        4   Subchunk2Size    == NumSamples * NumChannels * BitsPerSample/8
    This is the number of bytes in the data.
    You can also think of this as the size
    of the read of the subchunk following this
    number.
    44        *   Data             The actual sound data.


    The thing that makes reading wav files tricky is that java has no unsigned types.  This means that the
    binary data can't just be read and cast appropriately.  Also, we have to use larger types
    than are normally necessary.

    In many languages including java, an integer is represented by 4 bytes.  The issue here is
    that in most languages, integers can be signed or unsigned, and in wav files the  integers
    are unsigned.  So, to make sure that we can store the proper values, we have to use longs
    to hold integers, and integers to hold shorts.

    Then, we have to convert back when we want to save our wav data.

    It's complicated, but ultimately, it just results in a few extra functions at the bottom of
    this file.  Once you understand the issue, there is no reason to pay any more attention
    to it.

    ALSO:

    This code won't read ALL wav files.  This does not use to full specification.  It just uses
    a trimmed down version that most wav files adhere to.

     */

    ByteArrayOutputStream byteArrayOutputStream;
    AudioFormat audioFormat;
    TargetDataLine targetDataLine;
    AudioInputStream audioInputStream;
    SourceDataLine sourceDataLine;
    float frequency = 8000.0F;  //8000,11025,16000,22050,44100
    int samplesize = 16;
    private String myPath;
    private long myChunkSize;
    private long mySubChunk1Size;
    private int myFormat;
    private long myChannels;
    private long mySampleRate;
    private long myByteRate;
    private int myBlockAlign;
    private int myBitsPerSample;
    private long myDataSize;
    // I made this public so that you can toss whatever you want in here
    // maybe a recorded buffer, maybe just whatever you want
    public byte[] myData;



    public Wav()
    {
        myPath = "";
    }

    // constructor takes a wav path
    public Wav(String tmpPath) {
        myPath = tmpPath;
    }


    // get/set for the Path property
    public String getPath()
    {
        return myPath;
    }

    public void setPath(String newPath)
    {
        myPath = newPath;
    }

    // read a wav file into this class
    public boolean read() {
        DataInputStream inFile = null;
        myData = null;
        byte[] tmpLong = new byte[4];
        byte[] tmpInt = new byte[2];

        try {
            inFile = new DataInputStream(new FileInputStream(myPath));

            //System.out.println("Reading wav file...\n"); // for debugging only

            String chunkID = "" + (char) inFile.readByte() + (char) inFile.readByte() + (char) inFile.readByte() + (char) inFile.readByte();

            inFile.read(tmpLong); // read the ChunkSize
            myChunkSize = byteArrayToLong(tmpLong);

            String format = "" + (char) inFile.readByte() + (char) inFile.readByte() + (char) inFile.readByte() + (char) inFile.readByte();

            // print what we've read so far
            //System.out.println("chunkID:" + chunkID + " chunk1Size:" + myChunkSize + " format:" + format); // for debugging only



            String subChunk1ID = "" + (char) inFile.readByte() + (char) inFile.readByte() + (char) inFile.readByte() + (char) inFile.readByte();

            inFile.read(tmpLong); // read the SubChunk1Size
            mySubChunk1Size = byteArrayToLong(tmpLong);

            inFile.read(tmpInt); // read the audio format.  This should be 1 for PCM
            myFormat = byteArrayToInt(tmpInt);

            inFile.read(tmpInt); // read the # of channels (1 or 2)
            myChannels = byteArrayToInt(tmpInt);

            inFile.read(tmpLong); // read the samplerate
            mySampleRate = byteArrayToLong(tmpLong);

            inFile.read(tmpLong); // read the byterate
            myByteRate = byteArrayToLong(tmpLong);

            inFile.read(tmpInt); // read the blockalign
            myBlockAlign = byteArrayToInt(tmpInt);

            inFile.read(tmpInt); // read the bitspersample
            myBitsPerSample = byteArrayToInt(tmpInt);

            // print what we've read so far
            //System.out.println("SubChunk1ID:" + subChunk1ID + " SubChunk1Size:" + mySubChunk1Size + " AudioFormat:" + myFormat + " Channels:" + myChannels + " SampleRate:" + mySampleRate);


            // read the data chunk header - reading this IS necessary, because not all wav files will have the data chunk here - for now, we're just assuming that the data chunk is here
            String dataChunkID = "" + (char) inFile.readByte() + (char) inFile.readByte() + (char) inFile.readByte() + (char) inFile.readByte();

            inFile.read(tmpLong); // read the size of the data
            myDataSize = byteArrayToLong(tmpLong);


            // read the data chunk
            myData = new byte[(int) myDataSize];
            inFile.read(myData);

            // close the input stream
            inFile.close();
        } catch (Exception e) {
            return false;
        }

        return true; // this should probably be something more descriptive
    }

    // write out the wav file
    public boolean save() {
        try {
            DataOutputStream outFile = new DataOutputStream(new FileOutputStream(myPath + "temp"));

            // write the wav file per the wav file format
            outFile.writeBytes("RIFF");                 // 00 - RIFF
            outFile.write(intToByteArray((int) myChunkSize), 0, 4);     // 04 - how big is the rest of this file?
            outFile.writeBytes("WAVE");                 // 08 - WAVE
            outFile.writeBytes("fmt ");                 // 12 - fmt
            outFile.write(intToByteArray((int) mySubChunk1Size), 0, 4); // 16 - size of this chunk
            outFile.write(shortToByteArray((short) myFormat), 0, 2);        // 20 - what is the audio format? 1 for PCM = Pulse Code Modulation
            outFile.write(shortToByteArray((short) myChannels), 0, 2);  // 22 - mono or stereo? 1 or 2?  (or 5 or ???)
            outFile.write(intToByteArray((int) mySampleRate), 0, 4);        // 24 - samples per second (numbers per second)
            outFile.write(intToByteArray((int) myByteRate), 0, 4);      // 28 - bytes per second
            outFile.write(shortToByteArray((short) myBlockAlign), 0, 2);    // 32 - # of bytes in one sample, for all channels
            outFile.write(shortToByteArray((short) myBitsPerSample), 0, 2); // 34 - how many bits in a sample(number)?  usually 16 or 24
            outFile.writeBytes("data");                 // 36 - data
            outFile.write(intToByteArray((int) myDataSize), 0, 4);      // 40 - how big is this data chunk
            outFile.write(myData);                      // 44 - the actual data itself - just a long string of numbers
        } catch (Exception e) {
            System.out.println(e.getMessage());
            return false;
        }

        return true;
    }

    // return a printable summary of the wav file
    public String getSummary() {
        //String newline = System.getProperty("line.separator");
        String newline = "
"; String summary = "Format: " + myFormat + newline + "Channels: " + myChannels + newline + "SampleRate: " + mySampleRate + newline + "ByteRate: " + myByteRate + newline + "BlockAlign: " + myBlockAlign + newline + "BitsPerSample: " + myBitsPerSample + newline + "DataSize: " + myDataSize + ""; return summary; } public byte[] getBytes() { read(); return myData; } /** * Plays back audio stored in the byte array using an audio format given by * freq, sample rate, ect. * @param data The byte array to play */ public void playAudio(byte[] data) { try { byte audioData[] = data; //Get an input stream on the byte array containing the data InputStream byteArrayInputStream = new ByteArrayInputStream(audioData); AudioFormat audioFormat = getAudioFormat(); audioInputStream = new AudioInputStream(byteArrayInputStream, audioFormat, audioData.length / audioFormat.getFrameSize()); DataLine.Info dataLineInfo = new DataLine.Info(SourceDataLine.class, audioFormat); sourceDataLine = (SourceDataLine) AudioSystem.getLine(dataLineInfo); sourceDataLine.open(audioFormat); sourceDataLine.start(); //Create a thread to play back the data and start it running. It will run \ //until all the data has been played back. Thread playThread = new Thread(new PlayThread()); playThread.start(); } catch (Exception e) { System.out.println(e); } } /** * This method creates and returns an AudioFormat object for a given set * of format parameters. If these parameters don't work well for * you, try some of the other allowable parameter values, which * are shown in comments following the declarations. * @return */ private AudioFormat getAudioFormat() { float sampleRate = frequency; //8000,11025,16000,22050,44100 int sampleSizeInBits = samplesize; //8,16 int channels = 1; //1,2 boolean signed = true; //true,false boolean bigEndian = false; //true,false //return new AudioFormat( AudioFormat.Encoding.PCM_SIGNED, 8000.0f, 8, 1, 1, //8000.0f, false ); return new AudioFormat(sampleRate, sampleSizeInBits, channels, signed, bigEndian); } public void playWav(String filePath) { try { AudioClip clip = (AudioClip) Applet.newAudioClip(new File(filePath).toURI().toURL()); clip.play(); } catch (Exception e) { Logger.getLogger(Wav.class.getName()).log(Level.SEVERE, null, e); } } // =========================== // CONVERT BYTES TO JAVA TYPES // =========================== // these two routines convert a byte array to a unsigned short public static int byteArrayToInt(byte[] b) { int start = 0; int low = b[start] & 0xff; int high = b[start + 1] & 0xff; return (int) (high > 8) & 0x000000FF); b[2] = (byte) ((i >> 16) & 0x000000FF); b[3] = (byte) ((i >> 24) & 0x000000FF); return b; } // convert a short to a byte array public static byte[] shortToByteArray(short data) { return new byte[]{(byte) (data & 0xff), (byte) ((data >>> 8) & 0xff)}; } /** * Inner class to play back the data that was saved */ class PlayThread extends Thread { byte tempBuffer[] = new byte[10000]; public void run() { try { int cnt; //Keep looping until the input // read method returns -1 for // empty stream. while ((cnt = audioInputStream.read(tempBuffer, 0, tempBuffer.length)) != -1) { if (cnt > 0) { //Write data to the internal // buffer of the data line // where it will be delivered // to the speaker. sourceDataLine.write(tempBuffer, 0, cnt); } } //Block and wait for internal // buffer of the data line to // empty. sourceDataLine.drain(); sourceDataLine.close(); } catch (Exception e) { System.out.println(e); System.exit(0); } } } }
Chobot answered 17/1, 2011 at 16:29 Comment(0)
B
2

Answering this in 2018. I have a similar situation and realized a glaring mistake I made. Your "format" parameters in the argument of the constructor aren't correct.

    AudioFormat format = new AudioFormat(Encoding.PCM_SIGNED, 44100, 16, 2, 2, 44100, 
false);

The fifth parameter (in your case, the second "2") represents the frame size. Frame size = Sample size * Channels. Because your bit depth is 16, your sample size is 2 bytes.

Sample size = 2

Channels = 2

Frame size = Sample Size * Channels = 4

So, your line of code should read

    AudioFormat format = new AudioFormat(Encoding.PCM_SIGNED, 44100, 16, 2, 4, 44100, 
false);

Also, have you tried using the FormatConversionProvider?

    javax.sound.sampled.spi.FormatConversionProvider

https://docs.oracle.com/javase/tutorial/sound/converters.html This tutorial helped me a bunch, but I believe it assumes you've already imported the aforementioned class.

I didn't see these solutions posted on this thread, but perhaps you already figured it out. At any rate, hope this helps!

Balladist answered 16/7, 2018 at 22:40 Comment(2)
In the past 7 years I did at some point manage to figure it out. I have no idea what I was even doing, and haven't even written Java code for about 5 years. Thanks for your insight though!Xl
Haha, I figured, but maybe this answer will help someone searching for it in the future. I'm glad you figured it out. This stuff can be such a headache. I'm still quite the baby at programming, but I'd like to think I'm getting somewhere.Balladist

© 2022 - 2024 — McMap. All rights reserved.