Keyword detection method in voice signal, device, terminal and storage medium

Abstract

The invention discloses a keyword detection method in a voice signal, a device, a terminal and a storage medium, and belongs to the terminal technical field; the method comprises the following steps:obtaining a posterior probability of each target character that serves as a keyword in the voice signal in an acquisition period; obtaining the confidences of at least two target character combinations according to the posterior probability of each target character; determining the voice signal contains the keyword when the confidences of at least two target character combinations respectively satisfy preset conditions. The method obtains the confidences of at least two target character combinations, and determines the voice signal contains the keyword when the confidence of each target character combination satisfies preset conditions; the method obtains the confidence according to the posterior probability of each target character, and detects according to the confidence obtained by theposterior probability of certain target character, thus preventing wrong detection, and providing a more accurate detection result.

Description

technical field [0001] The present invention relates to the technical field of terminals, in particular to a method, device, terminal and storage medium for detecting keywords in voice signals. Background technique [0002] In modern life, smart devices such as smartphones, smart TVs, smart speakers, and smart switches have been widely used. Generally, smart devices have multiple functions, including voice signal collection and processing of real-time or non-real-time voice signals. Based on these functions, smart devices can execute according to keywords detected from voice signals without manual operation by users. Operate accordingly. Since the detection result of the keyword directly affects the accuracy of the operation performed by the smart device, how to detect the keyword in the voice signal becomes the key to improving the human-computer interaction experience. [0003] At present, the detection method of the keyword in the voice signal is: collect the voice sign...

AI Technical Summary

Problems solved by technology

[0004] However, in actual use scenarios, when the speech signal contains many words, there may be keywords that are not included in the speech signal, but the confidence obtained by accumulating the posterior probabilities of multiple words in the speech signal as keywords is less. is greater than the threshold, at this time it will be mistaken for the keyword detected from the speech signal, resulting in false detection

For example, if the keyword is "Ni Hao Xiao Ting", when the speech signal contains words that match the pronunciation of any three keywords in "Ni Hao Xiao Ting", the posterior probabilities of the three words as keywords are accumulated to obtain The confidence level of may be greater than the threshold, resulting in false detection

Images