Research On Speech Recognition Algorithm Of Hypersphere Cover Based On Hilbert Space

Speech recognition is a link of the voice chain, the ultimate goal of which is to make the computer understand whatever any person says and any content of the speech. Speech recognition technology as a cross discipline, has been very widely applied in household electronics, intellectual toys, the database voice inquiry of the business system, voice control in the industrial production sector, telephone and telecommunication systems automatic dialing field,etc. Although speech recognition technology has made some achievements, but due to the diversity and complexity of the voice signal, the current speech recognition efficiency remains to be improved, so the development of efficient speech recognition model and algorithm has become an important topic in the study of speech recognition. In this paper, the speech recognition technology in the preprocessing, feature extraction and pattern recognition of speech signal are studied in detail, the main contents of this paper are:In order to study the parameters that affect the speech recognition, we did the related feature extraction experiments. We found that the key of the speech recognition feature is the time-frequency domain feature by analyzing of the existing time-domain characteristics, frequency domain features and time- frequency domain features. By simplifying the spectrogram, we get a new time-frequency characteristics – zero-spectrum. A large number of experiments show that the zero-crossing spectroscopy is a simple characteristic function which has high recognition efficiency.The existing speech recognition model has high recognition precision but its computation is much complicated. In order to solve this problem, we proposed a new speech recognition model with high recognition precision and low computation complexity called hyper-sphere cover recognition algorithm based on Hilbert space. Through repeated experiments, we can know that the algorithm is lower than the traditional speech recognition algorithm, and the accuracy of speech recognition is not lower than the traditional speech recognition algorithm.In this paper, a new feature extraction method– zero-spectrum is proposed, and the recognition algorithm is hyper-sphere cover recognition algorithm based on Hilbert space. The simulation results show that the algorithm is fast and the recognition rate is high.The research of this paper is mainly used in low speed embedded system, the current speech recognition algorithm is relatively complex, resulting in a high price of voice chip, in the future we will continue to study voice recognition in real-time, accurate and speech recognition system in the direction of low price, so as to make speech recognition have more application value, especially in the Internet of things and Intelligent House System.