انجمن کامپیوتر ایران

One of the most important phases of speaker indexing is speaker clustering which aims to find the number of speakers in a speech document and merge the speech segments corresponding to a single speaker. The most critical source of problem in speaker clustering is the speech segments duration which may be so short that proper segment modeling becomes hard to achieve. An alternative suggestion in these situations is to adapt global models with new data instead of building the speaker models from the ground. In this paper we investigate two adaptation techniques in eigen-voice space for improving clustering performance especially for shorter speech utterances. These techniques were embedded in a clustering framework and evaluated on a set of domestic conversational speech. We have also compared the proposed methods with some other known techniques. The experiments show a considerable improvement in speaker clustering performance.

Reversible Cellular Automata is applicable in cryptographic functions. A reversible CA can be obtained by employing reversible rules. In this paper, we propose 65280 two-dimensional reversible CA rules by compounding one-dimensional reversible rules. We produce these rules by alternately applying onedimensional CA rule f1 to the rows and applying another rule f2 to the columns of the configuration matrix of twodimensional CA. In addition to describing these rules, we consider the required cryptographic properties such as completeness, strict avalanche criteria, non-linearity, and differential-profile flatness for our proposed approach. According to the obtained results, 9463 rules are specified as appropriate rules for the purpose of applying in cryptographic functions. Only 510 rules do not satisfy any of the mentioned cryptographic properties and the other remain rules satisfy some of the properties.