DPCM
From Just Solve the File Format Problem
(Difference between revisions)
m (link) |
|||
| Line 3: | Line 3: | ||
|subcat=Audio and Music | |subcat=Audio and Music | ||
}} | }} | ||
| − | '''Differential Pulse Code Modulation''' or '''DPCM''' is a method for encoding audio. It is a variant of pulse code modulation that quantizes and transmits the difference between consecutive samples rather than the sample values themselves. The decoder can then reconstruct the signal by adding consecutive DPCM samples. | + | '''Differential Pulse Code Modulation''' or '''DPCM''' is a method for encoding audio. It is a variant of [[pulse code modulation]] that quantizes and transmits the difference between consecutive samples rather than the sample values themselves. The decoder can then reconstruct the signal by adding consecutive DPCM samples. |
In practice, the naive method of quantizing after the difference filter can suffer from error accumulation. To account for this, DPCM encoders contain a model of the decoder, taking the difference with the decoder's reconstructed output rather than the actual last sample of the input audio. | In practice, the naive method of quantizing after the difference filter can suffer from error accumulation. To account for this, DPCM encoders contain a model of the decoder, taking the difference with the decoder's reconstructed output rather than the actual last sample of the input audio. | ||
DPCM generally achieves better quality for a given bitrate than PCM. The quantization method need not be linear, and can be [[µ-law]] or [[A-law]]. | DPCM generally achieves better quality for a given bitrate than PCM. The quantization method need not be linear, and can be [[µ-law]] or [[A-law]]. | ||
Latest revision as of 01:53, 7 June 2017
Differential Pulse Code Modulation or DPCM is a method for encoding audio. It is a variant of pulse code modulation that quantizes and transmits the difference between consecutive samples rather than the sample values themselves. The decoder can then reconstruct the signal by adding consecutive DPCM samples.
In practice, the naive method of quantizing after the difference filter can suffer from error accumulation. To account for this, DPCM encoders contain a model of the decoder, taking the difference with the decoder's reconstructed output rather than the actual last sample of the input audio.
DPCM generally achieves better quality for a given bitrate than PCM. The quantization method need not be linear, and can be µ-law or A-law.
