Spectral analysis of audio signal

Audio information is one of the largest part of all the information that a person can perceive through different organs of sense. Most often, people use speech to communicate with other people. But speech does not take the whole range of sound frequencies. In fact, sound information should be protected like any other information. And in order to understand what is needed to protect information, mechanisms for analyzing the spectrum of audio signal can help.Human speech takes approximately frequencies from 100 Hz to 5000 Hz, the range is individual for each person and may vary depending on the vocal cords. Telephony uses a frequency band from 300 Hz to 3400 Hz, which is enough to make out human speech transmitted over the transmission line. Here or in similar audio transmission systems, a spectrum analyzer can be used to determine the desired frequencies. Unnecessary frequencies can be eliminated, so you can get rid of unwanted noise that could distort the transmitted information.In the age of information technology, it should be understood that most electronic devices that receive, process, and transmit audio information work with an already converted signal that is not mechanical. In this regard, the same mechanism is applied to all signals to obtain their spectrum, only a device that can perform digitization of the signal is needed.Digitization involves two sub-processes: time sampling and level quantization. Time sampling is the representation of a signal by a discrete set of its values at different points in its time. For this subprocess, it is important to select the sampling step. The paper presents a mechanism for analyzing the spectrum of an audio signal. As an example, the reaction of the software package to the 12 KHz signal is shown. The software package allows changing the bandwidth within 20 KHz, which is a boundary indicator of human hearing.