基于百花知识图谱的语音识别智能问答系统构建OA

This paper takes the knowledge of flowers as an example to design and construct the knowledge graph,and builds a full-sequence convolutional neural network speech recognition model combined with Transformer.Based on the knowledge graph and speech recognition technology,the intelligent question and answering system of flower planting information is realized.Firstly,the planting information of flowers is obtained by crawler technology,then the entity,relation and attribute are extracted based on the defined data structure to construct the knowledge graph of flowers.Then,the acoustic model is built by using the deep full se-quence convolutional neural network model,and the speech recognition model is constructed by combining with the strong language model Transformer.Finally,on the basis of the established knowledge graph and speech recognition model,the intelligent question and answering system of flower knowledge is designed to locate the questions raised by users and provide users with personalized in-formation service through interface interaction.