Study on Consecutive Interpretation Automatic Scoring Model Based on Neural Network Algorithm

With the continuous development of artificial intelligence and the continuous improvement of hardware performance, the era of intelligence is coming rapidly. Using AI technology to help people deal with various practical problems has become increasingly important and urgent. Among them, the field of alternate interpretation has become a deep water area that artificial intelligence technology needs to conquer. In oral English teaching and automatic scoring, in recent years, reading questions, with questions to answer the automatic scoring system has reached the practical level, and reply, interpretation and a given range of automatic scoring method research is in the initial stage, related research results are less. Interpretation test is a comprehensive test of foreign language application ability, including foreign language thinking ability and language organization ability. Research and development of an effective Chinese-English interpretation automatic scoring system can not only provide students with an interpretation practice platform, but also assist teachers in teaching and release the pressure of teachers on teaching and marking. The paper provides the feasibility test of real test data in a provincial self-study test; using the literature research method, investigation method and quantitative analysis method, the average consensus rate of automatic score model results is 77%, which ensures the effectiveness and reliability of automatic score model based on neural network algorithm and can be fully utilized in the future.