摘要: For steganography over speech frames, least-significant-bit (LSB) approach has been one of the most important alternatives. However, its embedded capacity is quite limited, due to low redundancy characteristic bitrate speech. Thus, we suggest a novel least significant digit (LSD) method in this paper, which makes full use frame bits hide secret messages and provides larger embedding than LSB method. The LSD exploits multiple adjacent states parameters, are produced by modifications (e.g. +1, −1, +2, −2) encoded as LSDs using multi-ary numeration system. Beyond providing considerable capacity, further extends key space for key-based steganography, can enhance security covert communication. implemented evaluated with G.723.1 codec speeches. experiment results show that, compared previous method, increases around 30% induces less distortion given same capacity.