A method for finding the average spectrum of speech, comprising sound recording of speech material by speakers in a soundproof acoustic chamber by sound recording equipment with subsequent mixing of the sound recording; establishing a frequency range for further analysis of the frequency dependence of the spectral amplitude composition of the recorded speech material, characterized in that to find the average spectrum of speech using precision equipment: measuring non-directional condenser microphone, sound card, personal computer (PC) and
The method according to claim 1, characterized in that the speech material is recorded using speech material of at least three functional styles of speech, such as artistic, scientific, office-business, etc.
Method according to claim 1, characterized in that the speech material is recorded using a text containing at least 2500 characters excluding spaces.
Method according to claim 1, characterized in that the frequency range for further analysis of the amplitude composition of the speech material is set in the range from 100 Hz to 14-15 kHz.
Method according to claims. 1, 2, characterized in that the additional processing of the speech signal is carried out in the form of compression and equalization.
Method according to claims. 2, 3, characterized in that the identification of speech is determined by characteristic regions of the spectrum, which are found by processing the audio signal of an array of audiobooks.
Comments are closed