Data Compression

To compress something is to make it smaller. In terms of digital audio data, this refers to shrinking the file size and data flow of digital audio required to represent the audio signal. Data compression is used to maximize the fidelity of the product within the data restrictions of the playback medium.

In the realm of digital audio, numbers represent the audio signal. The more numbers used to describe the audio, the better the representation of the original sound. Better fidelity can be achieved in part by using higher sample rates and longer word lengths, i.e. more numbers, bigger files, and a larger flow of data to render the audio signal.

The professional world has accommodated the growth of file size and data flow by improving digital processing and developing cheap storage; the need for data compression arises in the final delivery to the consumer. There are lower limits of data flow in DVD, broadcast, and theatrical delivery systems. Compression solutions have been created, most notably by Dolby® Labs and DTS™, to significantly reduce the data flow required to represent the audio while maintaining a high degree of the fidelity inherent in the original, larger digital audio files. Data compression, (Dolby Digital AC-3, Dolby E, and DTS), is a separate process that utilizes either hardware in a real time conversion or batch processing with software.

Dolby and DTS hardware encoders.

A DTS audio compression software interface.