English translation robot pronunciation error correction method based on semantic matching

In order to improve the pronunciation accuracy of English translation robots, a semantic matching based pronunciation correction method for English translation robots is proposed. Use digital filters to pre emphasise the collected pronunciation signals to improve the accuracy and clarity of the pronunciation signals. Then, by calculating the semantic weight of the pronunciation signal and limiting the vector direction based on its weighted vector of semantic features, the pronunciation semantics of the English translation robot are matched. In order to further improve accuracy, the method of state separation can be used when processing feature quantities of pronunciation signals, and pronunciation error correction can be achieved through feature fusion. The experimental results show that this method is less affected by noise and can improve the reliability of robot pronunciation correction.