Project description
TranslatedTake at least the area of hearing. 2.1 (your work). Using regression analysis, approximate the lower abs curve. The hearing threshold, i.e. You get an analytic expression of the audible curve -- that's one little program in Mathcad.
The ADC you have so well painted in your work. Using the BPF (fast Fourier transform), a continuous audio signal is converted from the time domain to the frequency domain. Frequency components are obtained, which are compared with the obtained analytical expression of the audible curve.
And if the components in terms of level remained under this curve (Fig. 2 of your work), then they are not transmitted. Thus, the number of frequency components required for transmission can be reduced, i.e., the number of frequencies required for transmission can be reduced. Perform audio compression.
Programs must act as a filter (psychoacoustic model see ) Drawing. The components above the audible curve remain and are transmitted, and below are discarded.
This process should be modeled in MATLAB+Simulink. At the output, you should get a program that we feed an audio signal, listen to the compressed signal and count how many components are left, i.e., the program is designed to be used for the production of a sound signal. compression ratio.