Man-machine conversation method and system

Abstract

The invention discloses a man-machine conversation method and system which are applied to screen intelligent terminal equipment. The method comprises the following steps: starting a full-duplex conversation mode after detecting the conversation starting operation of a user; performing voice recognition on the detected current user statement, determining reply content corresponding to the current user statement according to an obtained voice recognition result, and presenting the reply content to the user; and if a new user statement is detected before the reply content corresponding to the current user statement is determined and presented to the user, determining the new reply content responding to the current user according to the current user statement and the new user statement. The full duplex dialogue mode is adopted, the new user statement can be detected in real time while the user statement is responded, and the new reply content responding to the current user is determined according to the current user statement and the new user statement, so that the reply content can be determined more accurately and efficiently by synthesizing the context at the same time.

Description

technical field [0001] The present invention relates to the technical field of man-machine dialogue, and in particular, to a man-machine dialogue method and system. Background technique [0002] At present, voice assistants running on smart TVs or OTT TV boxes on the market generally use half-duplex interaction for human-computer interaction. E.g: [0003] 1. Use the pickup board to add a wake-up word. You need to use the wake-up word for every conversation. You can interact after waking up. Wait for the VAD (voice detection) detection to end and stop picking up sounds. [0004] 2. Using the voice remote control, when the user presses the voice key, the dialogue starts and enters the voice pickup state. When the user lifts the voice key, the voice pickup stops and the dialogue output is output. [0005] Half-duplex interaction can satisfy most simple operations, such as increasing the volume, changing TV channels, switching songs up and down, etc. But when dealing with co...

AI Technical Summary

Problems solved by technology

[0013] 2. The effect of instant interruption (termination of the current dialogue process) cannot be achieved, and the dialogue needs to be interrupted or wait for the end of the dialogue to enter the next round of dialogue;

[0014] 3. In scenarios where message referencing is required, it is impossible to infer user intent based on the context after cross-domain implementation;

[0015] 4. The operation mode is fixed, and the interactive experience is not good

This solution exists, but the number of wake-up words is limited, the wake-up effect is not good, it is easy to wake up or wake up by mistake, and cannot deal with cross-domain scenarios and other problems

Images