Fan Yanhong

Adaptive acoustic masking based on spectral envelope

Acoustic masking can be used to protect the voice information from being eavesdropped. This paper presents a method to generate masking signal which is similar to white noise. It adjusts the white noise spectral energy by linear prediction of the masked speech spectral envelope in real time. The proposed method can accurately determine the frequency spectrum of the masking speech signal, and can adaptively adjust the masking noise volume by detecting masked speech energy in time domain. This algorithm was tested by using SII (Speech Intelligibility Index) defined in American National Standard ANSI S3.5. Experimental results show that this scheme provides better acoustic comfort and better masking capability compared with white masking noise. The algorithm can significantly improve the difficulty of stealing useful voice information by eavesdropper.

Design and implementation of Android-based speech storage system

With rapid development and wide application of the mobile intelligent terminal equipments, increasingly huge storage space is required. Therefore, it is particularly important to design a storage system based on the current mainstream operating system Android. In this paper, a high-quality speech storage system based on MELP speech coding algorithm is designed and implemented on Android system. The speech rate can be compressed to less than 0.5% with high speech quality level to provide more efficient speech transmission and storage for mobile intelligent terminals.

Acoustic masking based on time-reversed speech

Acoustic masking refers to the effect that a sound can be suppressed by another. It can be used for speech privacy protection. According to human auditory characteristics, better acoustic masking can be obtained when the spectrum of the masking acoustic signal is close to the target speech. This paper presents an improved acoustic masking algorithm based on time-reversion of the speech to be masked. Time-reversed speech preserves the speakers pronunciation features but is meaningless in semantics. The proposed method can significantly improve the security of speech information and the deciphering difficulty for eavesdropping. The experimental results show that the generated speech-like noise has a good masking effect and is comfortable for speakers.

Research on Anti-Laser-Eavesdropping Devices

Research of anti-laser-eavesdropping devices is very critical to the security of confidential meetings for the nation and enterprises. Based on the depth study of laser eavesdropping principle as well as the human hearing characteristics, an implementation of anti-laser-eavesdropping was proposed. The system is composed of the noise generators and vibration transducers. Voice signals before and after eavesdropping were compared, the results of which show that the anti-laser-eavesdropping devices could produce noise signal which could mask the voice signal and successfully block the laser eavesdropping channels.