Text Independent Open-Set Cell phone Identification

Abstract

This paper discusses the application of speech signals that convey various pieces of information such as the identity of its speaker, the language spoken, and the linguistic information about the text being spoken etc. The rapid developments in technologies related to cell-phones have resulted in their much broader usage than mere talking devices used for making and receiving phone calls. User-generated audio recordings from cell phones can be very helpful in a number of forensic applications. This thesis proposes a novel system for open-set cell-phone identification from speech samples recorded using the cell-phone. The proposed system uses different features based on original speech recordings and classifies them using sequential minimal optimization (SMO) based Support vector machine (SVM) and Vector Quantization (VQ). The performance of the proposed system is tested on a customised databases extracted from pre-recorded speech content of twenty-two cell phones of different manufacturers. Closed-set cell-phone recognition systems abound, and the overwhelming majority of research in cell-phone recognition in the past has been limited to this task. A realistically viable system must be capable of dealing with the open-set task. This effort attacks the open-set task, identifying the best features to use, and proposes the use of a fuzzy classifier followed by hypothesis testing as a model for text-independent, open-set cell-phone recognition.