An Analog Implementation of the Psychoacoustic Model Used in Audio Compression Algorithms on an FPAA
Loading...
Author(s)
Liu, Lenno Dawei
Advisor(s)
Hasler, Jennifer O.
Editor(s)
Collections
Supplementary to:
Permanent Link
Abstract
Lossy audio compression takes advantage of how humans perceive sound to achieve high compression ratios. This is done by analyzing the audio to determine what is audible to the human ear using a psychoacoustic model, and preferentially allocating more bits to encode the audible portions while cutting down on the bits used to encode the masked or inaudible portions. This method allows for an approximately tenfold reduction in the memory required without a perceptible loss in quality, and is used everywhere in the modern world from VoIP calls to music storage and streaming services. All current audio compression algorithms work on a digital in, digital out basis converting from a pulse-code modulated (PCM) signal to a bitstream output. This work aims to provide a basis to create a mixed signal encoder that takes an analog input for the same digital output, by implementing an analog psychoacoustic model on a field-programmable analog array (FPAA).
Sponsor
Date
2023-05-05
Extent
Resource Type
Text
Resource Subtype
Thesis