Speech Recognition Systems: Disadvantages and Limitations

Speech is one of the easiest ways allowing communication between people and machines. Speech recognition technology makes everyday life easier: it is widely used in mobile phones, computers, tablets, cars, etc. However, the quality of automatic speech recognition is affected by many factors, therefore, so much effort is put into improving the performance of speech recognition systems. The aim of this paper is to present the current state of development of speech recognition systems and to examine their drawbacks and limitations. The paper discusses the current classification, construction and functioning of speech recognition systems, which gives an insight into the speech-to-text software implemented so far. The analysis of disadvantages and limitations of speech recognition systems has allowed identifying the weak points of these systems. Problems that are to be solved in the near future indicate the direction of further development of speech recognition systems.