Automatic speech recognition systems: challenges and recent implementation trends

Speech recognition is one of the next generation technologies for human–computer interaction. Speech recognition has been researched since the late 1950s but due to its computational complexity and limited computing capabilities of the last few decades, its progress has been impeded. In laboratory settings automatic speech recognition systems (ASR) have achieved high levels of recognition accuracies, which tend to degrade in real world environments. This paper analyses the basics of the speech recognition system. Major problems faced by ASR in real world environments have been discussed with major focus on the techniques used in the development of noise robust ASR. Throughout the years there have been different implementation mediums for ASR but Field Programmable Gate Arrays (FPGAs) seems to provide a unique advantage for the implementation of Digital Signal Processing (DSP) systems and by extension ASR systems.