Therefore, a need has been felt for the design of a dedicated hardware for random number generator that generates one random number per cycle so that text-to speech conversion is done in real time.
For parametric speech synthesis application, a random number generator is required to produce noise samples. Most C libraries have a pair of library routines for initializing, and then generating random numbers. Random number generator is required extensively by many applications like cryptography, simulation, numerical analysis, text-to-speech etc. This paper discusses in detail the hardware implementation of several RNGs and their characteristics. FPGAs make it possible to implement complex systems, such as numerical calculations, genetic programs, simulation algorithms etc., at hardware level. Although a majority of random number generators have been implemented in software level, increasing demand exists for hardware implementation due to the advent of faster and high density Field Programmable Gate Arrays (FPGA). True random number generators are slow and expensive for many applications while pseudo random number generators (RNG) suffice for most applications. Random numbers are used in a wide variety of applications.
