G726

java.lang.Object
- org.zoolu.sound.codec.G726

Direct Known Subclasses:

G726_24, G726_32, G726_40
```
public abstract class G726
extends java.lang.Object
```
Common routines for G.721 and G.723 conversions.
This implementation is based on the ANSI-C language reference implementations of the CCITT (International Telegraph and Telephone Consultative Committee) G.711, G.721 and G.723 voice compressions, provided by Sun Microsystems, Inc.
Acknowledgement to Sun Microsystems, Inc. for having released the original ANSI-C source code to the public domain.

Field Summary

Fields
Modifier and Type	Field and Description
`static int`	`AUDIO_ENCODING_ALAW` ISDN A-law
`static int`	`AUDIO_ENCODING_LINEAR` PCM 2's-complement (0-center)
`static int`	`AUDIO_ENCODING_ULAW` ISDN u-law

Constructor Summary

Constructors
Constructor and Description

G726(javax.sound.sampled.AudioFormat.Encoding encoding)
Creates a new G726 processor, that can be used to encode from or decode do PCM audio data.

Constructors
Constructor and Description
`G726(javax.sound.sampled.AudioFormat.Encoding encoding)` Creates a new G726 processor, that can be used to encode from or decode do PCM audio data.

Method Summary

All Methods Static Methods Instance Methods Abstract Methods Concrete Methods
Modifier and Type	Method and Description
`abstract int`	`decode(byte[] in_buff, int in_offset, int in_len, int out_coding, byte[] out_buff, int out_offset)` Decodes the input chunk in_buff of G726 encoded data and returns the linear PCM, A-law or u-law chunk into out_buff.
`abstract int`	`decode(int i, int out_coding)` Decodes a 4-bit code of G.72x encoded data of i and returns the resulting linear PCM, A-law or u-law value.
`abstract int`	`encode(byte[] in_buff, int in_offset, int in_len, int in_coding, byte[] out_buff, int out_offset)` Encodes the input chunk in_buff of linear PCM, A-law or u-law data and returns the G726 encoded chuck into out_buff.
`abstract int`	`encode(int sl, int in_coding)` Encodes the input vale of linear PCM, A-law or u-law data sl and returns the resulting code. -1 is returned for unknown input coding value.
`javax.sound.sampled.AudioFormat.Encoding`	`getEncoding()` Gets the G726 encoding type.
`protected static int`	`quantize(int d, int y, int[] table, int size)` Given a raw sample, 'd', of the difference signal and a quantization step size scale factor, 'y', this routine returns the ADPCM codeword to which that sample gets quantized.
`protected static int`	`reconstruct(int sign, int dqln, int y)` Returns reconstructed difference signal 'dq' obtained from codeword 'i' and quantization step size scale factor 'y'.
`static int`	`signedIntLittleEndian(byte hi_b, byte lo_b)` Converts 2 little-endian-bytes into a signed int.
`protected static int`	`tandem_adjust_alaw(int sr, int se, int y, int i, int sign, int[] qtab)` At the end of ADPCM decoding, it simulates an encoder which may be receiving the output of this decoder as a tandem process.
`protected static int`	`tandem_adjust_ulaw(int sr, int se, int y, int i, int sign, int[] qtab)`
`protected static int`	`unsignedInt(byte b)` Converts a byte into an unsigned int.
`static int`	`unsignedIntLittleEndian(byte hi_b, byte lo_b)` Converts 2 little-endian-bytes into an unsigned int.
`protected static void`	`update(int code_size, int y, int wi, int fi, int dq, int sr, int dqsez, org.zoolu.sound.codec.G726State state)` updates the state variables for each output code

Methods inherited from class java.lang.Object
clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

- Field Detail
  - AUDIO_ENCODING_ULAW
```
public static final int AUDIO_ENCODING_ULAW
```
    ISDN u-law
    
    See Also:
    
    Constant Field Values
  - AUDIO_ENCODING_ALAW
```
public static final int AUDIO_ENCODING_ALAW
```
    ISDN A-law
    
    See Also:
    
    Constant Field Values
  - AUDIO_ENCODING_LINEAR
```
public static final int AUDIO_ENCODING_LINEAR
```
    PCM 2's-complement (0-center)
    
    See Also:
    
    Constant Field Values
- Constructor Detail
  - G726
```
public G726(javax.sound.sampled.AudioFormat.Encoding encoding)
```
    Creates a new G726 processor, that can be used to encode from or decode do PCM audio data.
- Method Detail
  - quantize
```
protected static int quantize(int d,
                              int y,
                              int[] table,
                              int size)
```
    Given a raw sample, 'd', of the difference signal and a quantization step size scale factor, 'y', this routine returns the ADPCM codeword to which that sample gets quantized. The step size scale factor division operation is done in the log base 2 domain as a subtraction.
    
    Parameters:
    
    d - - Raw difference signal sample
    
    y - - Step size multiplier
    
    table - - Quantization table
    
    size - - Table size of short integers
  - reconstruct
```
protected static int reconstruct(int sign,
                                 int dqln,
                                 int y)
```
    Returns reconstructed difference signal 'dq' obtained from codeword 'i' and quantization step size scale factor 'y'. Multiplication is performed in log base 2 domain as addition.
    
    Parameters:
    
    sign - - 0 for non-negative value
    
    dqln - - G.72x codeword
    
    y - - Step size multiplier
  - update
```
protected static void update(int code_size,
                             int y,
                             int wi,
                             int fi,
                             int dq,
                             int sr,
                             int dqsez,
                             org.zoolu.sound.codec.G726State state)
```
    updates the state variables for each output code
    
    Parameters:
    
    code_size - - distinguish 723_40 with others
    
    y - - quantizer step size
    
    wi - - scale factor multiplier
    
    fi - - for long/short term energies
    
    dq - - quantized prediction difference
    
    sr - - reconstructed signal
    
    dqsez - - difference from 2-pole predictor
    
    state - - coder state
  - tandem_adjust_alaw
```
protected static int tandem_adjust_alaw(int sr,
                                        int se,
                                        int y,
                                        int i,
                                        int sign,
                                        int[] qtab)
```
    At the end of ADPCM decoding, it simulates an encoder which may be receiving the output of this decoder as a tandem process. If the output of the simulated encoder differs from the input to this decoder, the decoder output is adjusted by one level of A-law or u-law codes.
    
    Parameters:
    
    sr - - decoder output linear PCM sample,
    
    se - - predictor estimate sample,
    
    y - - quantizer step size,
    
    i - - decoder input code,
    
    sign - - sign bit of code i
    
    Returns:
    
    adjusted A-law or u-law compressed sample.
  - tandem_adjust_ulaw
```
protected static int tandem_adjust_ulaw(int sr,
                                        int se,
                                        int y,
                                        int i,
                                        int sign,
                                        int[] qtab)
```
    Parameters:
    
    sr - - decoder output linear PCM sample
    
    se - - predictor estimate sample
    
    y - - quantizer step size
    
    i - - decoder input code
    
    sign -
    
    qtab -
  - unsignedInt
```
protected static int unsignedInt(byte b)
```
    Converts a byte into an unsigned int.
  - unsignedIntLittleEndian
```
public static int unsignedIntLittleEndian(byte hi_b,
                                          byte lo_b)
```
    Converts 2 little-endian-bytes into an unsigned int.
  - signedIntLittleEndian
```
public static int signedIntLittleEndian(byte hi_b,
                                        byte lo_b)
```
    Converts 2 little-endian-bytes into a signed int.
  - getEncoding
```
public javax.sound.sampled.AudioFormat.Encoding getEncoding()
```
    Gets the G726 encoding type.
  - encode
```
public abstract int encode(int sl,
                           int in_coding)
```
    Encodes the input vale of linear PCM, A-law or u-law data sl and returns the resulting code. -1 is returned for unknown input coding value.
  - encode
```
public abstract int encode(byte[] in_buff,
                           int in_offset,
                           int in_len,
                           int in_coding,
                           byte[] out_buff,
                           int out_offset)
```
    Encodes the input chunk in_buff of linear PCM, A-law or u-law data and returns the G726 encoded chuck into out_buff.
    It returns the actual size of the output data, or -1 in case of unknown in_coding value.
  - decode
```
public abstract int decode(int i,
                           int out_coding)
```
    Decodes a 4-bit code of G.72x encoded data of i and returns the resulting linear PCM, A-law or u-law value. return -1 for unknown out_coding value.
  - decode
```
public abstract int decode(byte[] in_buff,
                           int in_offset,
                           int in_len,
                           int out_coding,
                           byte[] out_buff,
                           int out_offset)
```
    Decodes the input chunk in_buff of G726 encoded data and returns the linear PCM, A-law or u-law chunk into out_buff.
    It returns the actual size of the output data, or -1 in case of unknown out_coding value.

Class G726

Field Summary

Constructor Summary

Method Summary

Methods inherited from class java.lang.Object

Field Detail

AUDIO_ENCODING_ULAW

AUDIO_ENCODING_ALAW

AUDIO_ENCODING_LINEAR

Constructor Detail

G726

Method Detail

quantize

reconstruct

update

tandem_adjust_alaw

tandem_adjust_ulaw

unsignedInt

unsignedIntLittleEndian

signedIntLittleEndian

getEncoding

encode

encode

decode

decode