Voice command arbitration method and device

Abstract

The embodiment of the invention provides a voice command arbitration method and device, and belongs to the technical field of computers. The voice command arbitration method comprises the steps: voicedata are obtained; the obtained voice data are recognized, and a voice command corresponding to the voice data is obtained; and whether one or more of a current interface, a current application program and an operating system need to process the voice command or not is inquired, and if yes, a callback function corresponding to the voice command is sent to one or more of the current interface, thecurrent application program and the operating system. According to the voice command arbitration method and device, through unified execution of recording, recognition and analysis of the voice data,and hierarchical execution of arbitration and distribution of the voice command, the reusability of a code is improved, the complexity of the code is also lowered, and the processing efficiency of the voice command is improved.

Description

technical field [0001] The present invention relates to the field of computer technology, and in particular to a voice command arbitration method and device. Background technique [0002] The methods of processing voice commands in the prior art are mainly divided into two categories. One is that each App that uses the voice recognition function performs a complete set of voice service processes such as recording, noise cancellation, recognition, analysis, and response. Although the App can also The above process is realized by integrating a third-party voice SDK (Software Development Kit, software development kit), but most of the logic is still encapsulated in the SDK. The code cannot be reused, resulting in a waste of code. [0003] The other type is that there is a unified voice service module at the system level, which can handle recording, noise reduction, recognition and analysis processes in a unified manner at the bottom layer of the system level, and then distribu...

AI Technical Summary

Problems solved by technology

[0002] The methods for processing voice commands in the prior art are mainly divided into two categories. One is that each App that uses the voice recognition function performs a complete set of voice service processes such as recording, noise elimination, recognition, analysis, and response. Although the App can also The above-mentioned process is realized by integrating a third-party voice SDK (Software Development Kit, software development kit), but most of the logic is still encapsulated in the SDK to achieve, the essence is the same as the App itself to execute the entire process, the same function The code cannot be reused, resulting in a waste of code

[0003] The other type is to have a unified voice service module at the system level, which can uniformly process recording, noise cancellation, recognition, and analysis processes at the bottom of the system level, and then distribute voice commands to the corresponding APP for execution through a command distribution center. However, the analysis and distribution of such voice commands are all concentrated in the voice service module, resulting in bloated module code, complex business, and frequent maintenance of the code

Images