Wake-up-free interaction method and device

Abstract

The invention discloses a wake-up-free interaction method and device, and the method comprises the steps: responding to a received effective voice signal of a user, and determining a detection interval based on the time period of the effective voice signal; if the effective pointing feature of the user is detected in the detection interval, judging whether the effective voice signal and the effective pointing feature correspond to an effective instruction or not; and if the corresponding effective instruction is judged, processing and feeding back the effective instruction. According to the scheme, the wake-up-free interaction is achieved by combining the pointing feature with voice interaction, the interaction experience of a user and intelligent voice equipment can be improved under the condition that the low wake-up error rate is ensured, the user interaction efficiency can be greatly improved especially in a frequent interaction scene, and meanwhile, the mode of the wake-up feature is used as the multi-mode input of the dialogue system, so that the application scene of the voice dialogue system is enriched.

Description

technical field [0001] The invention belongs to the field of voice recognition, in particular to a wake-up-free interaction method and device. Background technique [0002] In order to improve the accuracy of interaction and avoid misuse, current smart devices generally need to be woken up before they can perform voice interaction. At present, the wake-up technology is still based on voice wake-up. In order to avoid misuse, some technologies support registering a few commands as quick wake-up words, such as previous song and next song. However, introducing too many quick wake-up words will increase the chance of false wake-up. probability, so the use of quick wake-up words is relatively restrained. There are also some technologies that support the interaction without using the wake-up word after the face is recognized in a specific scene, but this will require the user's posture, and the face recognition must be successful before the operation can be performed. And due to ...

AI Technical Summary

Problems solved by technology

[0003] There are currently existing technical solutions related to wake-up free on the market: some solutions require an additional wireless headset device to calculate the distance, but even if the distance is very close, it does not mean that you are talking to the speaker; some solutions simply set Some wake-up-free words, such as the previous song and the next song, and this solution is mainly for the currently running application. Too many wake-up words will also lead to an increase in the false wake-up rate; some solutions are for the detection of eye sight Low precision, for example, when the eyes are looking at the keyboard, it is impossible to capture whether to look at H or G; some solutions are not used for waking up, but only for maintaining dialogue; some solutions have the defect that they cannot avoid waking up when there are many people

There is no wake-up-free solution that is more user-friendly in the prior art

Images