A sparse fuzzy c-means algorithm based on sparse clustering framework

Application research of artificial intelligence English audio translation system based on fuzzy algorithm

With the development of globalization, people’s demand for English audio interaction is increasing. In order to overcome the shortcomings of traditional translation methods in grammatical variables, such as semantic ambiguity, quantifier errors, low translation accuracy, improve the quality and speed of English translation, and get more accurate and speed guaranteed translation, this study proposes an artificial intelligence English audio translation cross language system based on fuzzy algorithm. In this experiment, the collected analog speech signal is converted into a digital speech signal, and then, the speech features are modeled and digitized, and the whole set of speech samples are integrated and modified to eliminate the interference caused by noise as far as possible. After that, the collected voice will be stored in the text format, and then the text will be translated to achieve English audio translation. The DNN-HMM speech recognition model and the traditional GMM-HMM speech recognition model are used to preprocess the original corpus, and the accuracy of the corpus processing is compared. After that, the accuracy and utilization of the fuzzy algorithm are evaluated between the first type TSK and the second type TSK. For speech synthesis in which the corpus lacks language, it is meaningful to explore the least amount of training data for the synthesis of acceptable speech. The experimental results show that the accuracy of the fuzzy algorithm is about 97.34%, and the utilization rate is about 98.14%. The accuracy rate of type 1 and type 2 algorithms are about 85.77% and 76.87% respectively, and the utilization rate is about 83.25% and 78.63% respectively. The fuzzy algorithm based artificial intelligence English audio translation cross language system is obviously better than the other two algorithms.