SPEAKER VOICE DISTORTION SYSTEM Russian patent published in 2010 - IPC G10L15/00 

FIELD: physics.

SUBSTANCE: system has a reference signal generator, a device for generating parametres for breaking the frequency band, integrators, a threshold level generating device, a power supply, a device for generating parametres for shifting positions of significant spectrum intervals, a digital recording device, a discretisation unit, a discrete Fourier transformation unit, main and two extra Fourier spectrum normalisation units, main and extra units for generating Fourier spectrum envelope, a unit for generating parametres of variation of the Fourier spectrum envelope, main and extra comparators, main and extra adders, a unit for determining maxima and corresponding arguments, a device for recording maximum values and corresponding arguments, a dynamic programming unit, a decision block, a unit for determining positions of significant spectrum intervals, a unit for determining shifted positions of significant spectrum intervals, an inverse Fourier transformation unit, a memory unit, a phase smoothing unit and a playback unit. Principle of operation of the system is based on using statistical signal processing methods and apparatus for generating spectral characteristics of a speaker voice. The system can be used to enable distortion of a speaker voice in accordance with given parametres, listening a distorted speaker voice, selection of the final version of distortion, as well as for real-time transmission of a distorted speaker voice to the output of an audio device with possibility of further transmission over communication channels.

EFFECT: low probability of voice recognition due to use of different versions of transforming spectral characteristics of the initial speaker voice.

7 dwg