Quick voice interaction method and system

Abstract

The invention discloses a quick voice interaction method and system. The method includes receiving voice input of a user in real time; performing real time voice identification on the voice input information and acquiring a voice identification result; performing real time semantic comprehension on the current voice identification result based on a pre-constructed semantic comprehension model andacquiring a model output result; determining whether a correct and complete semantic comprehension result is in presence or not according to the model output result; generating an interaction result according to the semantic comprehension result and sending the interaction result to the user as a feedback if the correct and complete semantic comprehension result is in presence; continuing real time voice identification on the voice input information if the correct and complete semantic comprehension result is not in presence. According to the invention, feedback delay of voice interaction canbe reduced and user experience can be improved.

Description

technical field [0001] The invention relates to the fields of voice signal processing and human-computer interaction, in particular to a fast voice interaction method and system. Background technique [0002] With the advancement of artificial intelligence technology, human-computer voice interaction has also made great progress, and various voice assistant APPs and human-computer interaction robots are booming. The process by which these voice assistants and bots interact with users is usually as follows: [0003] (1) Wake up by voice or setting operation; [0004] (2) Judge whether the user has finished speaking by detecting the end point of the voice; [0005] (3) After detecting that the user has finished speaking, the recognition result is passed to the semantic understanding engine for semantic understanding; [0006] (4) According to the result of semantic understanding, the system responds. [0007] In the above process, speech recognition and semantic understand...

AI Technical Summary

Problems solved by technology

Speech post-endpoint detection needs to set the silence duration, which is usually between 0.8s and 1.8s. Therefore, from the time when the user actually finishes speaking to the time when the system returns the interactive understanding result, regardless of the time-consuming semantic understanding engine, there is at least one Silence duration such a hard delay

And in a noisy environment, the detection accuracy of the backend point is not high, which will bring greater hard delay and affect the user experience

Images