Voice interaction method, electronic equipment and storage medium

Abstract

The invention discloses a voice interaction method, electronic equipment and a storage medium. The voice interaction method comprises the following steps: acquiring user voice data to perform voice recognition in real time to obtain a user voice request; under the condition that the complete user voice request is not received, complementing the user voice request acquired in real time based on the prefix tree to obtain a complementing result; processing the completion result to obtain a predicted voice instruction; and after the complete user voice request is received, if the completion result is the same as the received complete user voice request, completing voice interaction according to the predicted voice instruction. According to the method, instruction prediction closely associated with an individual can be constructed on the basis of the prefix tree for the received incomplete user voice request, automatic complementation of words strongly associated with the individual is realized, the technical effect that personalized words of different users can be recognized by thousands of people and thousands of faces is realized, the scheme is high in controllability, the requirement for required equipment is low, and the method is suitable for popularization and application. The realization is easy.

Description

technical field [0001] The present invention relates to the technical field of voice interaction, and in particular, to a voice interaction method, an electronic device and a storage medium. Background technique [0002] For text completion, similar services include prompt words for input methods, input prompts for search boxes, code completion prompts, and automatic text generation. First, conventional search algorithms can be used, but this requires storing a large amount of data in advance, and searching in a huge database, using the space-for-time method to improve efficiency and accuracy. Second, you can also use the popular pre-trained text generation models in recent years. Although the accuracy rate is high and the diversity is rich, the model parameters are huge, and it takes a long time to train texts in specific fields, and the time to generate text is also Longer, higher requirements for equipment, and greater time consumption. [0003] However, in the scenario...

AI Technical Summary

Problems solved by technology

Second, you can also use the popular pre-trained text generation model in recent years. Although the accuracy rate is high and the diversity is rich, the model parameters are huge. Longer, higher requirements for equipment, and more time spent

[0004] At the same time, each user has their own speaking habits. It is unscientific to use a large model to predict everyone's speech. Forcing the model to memorize and learn may be counterproductive, resulting in poor prediction of common speech and failure to recognize different User's personalized words, poor user experience

Images