Speech Recognition System Based on Duty Log of Distribution Network

In-depth analysis of the business needs of the power industry, through the speech database of specific professional terms, integrated with the existing machine learning technology, to form a fast and efficient interactive voice recognition system for distribution network duty log, has strong use value. In order to enrich the shortcomings of the existing research on speech recognition of the distribution network shift log, this paper discusses the function equation of the distribution network topology analysis algorithm, the shift log and the application of support vector machine in machine learning in speech recognition. This paper briefly discusses the development environment and sample data of the speech recognition system for the duty log of distribution network. In addition, the flow structure of the speech recognition system of the power grid shift log with support vector machine proposed in this paper is designed and discussed. Finally, the speech recognition system of the power distribution network shift log designed in this paper is applied to the four dispatching terms of command, reply, accident report and self-operation for experimental testing and training. The experimental data show that the recognition accuracy of support vector machine is 98.9% on average, which is better than the neural network recognition model. On the other hand, the recognition accuracy of accident reporting and self-operation dispatching terms reached 97.2% on average. Therefore, it is verified that the speech recognition system of distribution network duty log based on support vector machine has better performance effect.