An investigation into the effect of security on reliability and voice recognition system in a VoIP network

Today, communications technology is one of the most essential roles for our life. People had used many ways of transmissions such as internet, mobile phone and so on. Moreover, VoIP (Voice over Internet Protocol) is also a communications technology. In this paper, we focus on VoIP which transmits voice over packet switched networks such as LANs (Local Area Networks) or WANs (Wide Area Networks). Adding security to a VoIP system, the quality of service and performance of the system are at risk. The contribution of this system has two main parts. Firstly, it illustrates a key from Diffie-Hellman Key Exchange Algorithm. This key is used by AES Algorithm for encryption/decryption process to secure the signaling and voice traffic within a VoIP system. Secondly, it evaluates the performance of a VoIP system to develop speech recognition at receiver. In this system, Mfccs (Mel-Frequency Cepstral Coefficients) is utilized for feature extracting from the incoming voice at receiver. Moreover, Gaussian algorithm is also used for vector quantization in feature matching MFCC features to represent raw speech signal into compact but effective representation that is more stable and discriminative than the original signal. And then the breakdown of call setup is applied by Counting Process in Stochastic Processes. So, this system had captured and analyzed registration, calls setup, voice transmission packets and finally recognized voice or speech by using Mfccs in DSP (Digital Signal Processing) technique.