Abstract
DSD audio is a format, that represents a waveform as a series of 1s and 0s, rather than multibit samples how PCM audio does. Most lossy transform audio codecs, like MP3, AAC, Vorbis and Opus, encode PCM audio, and decode compressed audio back into the PCM format. If DSD audio has to be compressed using auch a codec, it first has to be converted to the PCM format. This conversion is said to introduce audible noise. Therefore it is desireable to have a lossy codec, that can compress DSD directly without resorting to PCM as an intermediate format. This disclosure describes a method to compress DSD audio in a lossy way, similar to MP3 or AAC. A time-domain to frequency-domain transform is applied directly to the 1-bit audio, and the coefficients for the ultrasonic frequencies are discarded.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
Schmidt, Simon, "Method for lossy compression of DSD Audio", Technical Disclosure Commons, (November 24, 2024)
https://www.tdcommons.org/dpubs_series/7591