Device for speech-like noise synthesis

Abstract

The feasibility study of using active means of speech information protection with flexible configuration based on wireless connection is justified. Speechlike noise generating device based on wireless connection for speech information protection is proposed. It is proposed to form two masking speech-like signals, one with a power equal to the power of the information signal and generated without taking into account the statistical features of the language, the second with significantly greater power and generated taking into account the statistical features of the language, which will increase the resistance of the protection to methods of statistical analysis and signal power analysis. A speech information protection device body modeled in the SolidWorks environment is proposed for subsequent 3D printing. PLA plastic was used in the printing process; a photo of the assembled speech information protection device is presented below in the text of the article. Testing of the developed device was carried out using the Salute Speech neural network.