Voice control device

Abstract

The invention provides a voice control device. Through an automatic speech recognition (ASR) system, sound signals collected by a sound collecting device are processed to obtain command signals, according to the command signals, voice control is achieved, and the voice control device comprises an audio pre-processing module, an analog-digital conversion module, an effective voice command generation module and an answer statement playing module; and according to the answer statement playing module, a controlled terminal detects and receives voice control commands in real time or regularly, whenthe voice control commands are detected, the controlled terminal determines whether or not the detected voice control commands correspond to answer statements. According to the voice control device,for application occasions of in-ear microphones, a series of optimizing methods are provided, noise components are effectively inhibited, and thus pure voice data is obtained and then pushed to a relative back-end service module, so that high-quality services are provided.

Description

technical field [0001] The technical field of voice control of the present invention mainly relates to a voice control device. Background technique [0002] Listening and speaking are the basic attributes of human communication. Correspondingly, smart wearable devices also require basic communication skills, being able to speak and listen. The hearing ability of smart wearable devices is accomplished by two kinds of input and output acoustic sensors. In the former, the sound is collected through the microphone input sensor; in the latter, the sound is played through the speaker output sensor. Generally speaking, the sound collected by an ordinary microphone does not contain only a single sound, but a mixture of human voice, music sound, and various environmental sounds (noise). The completion of many services provided by smart wearable devices (voice wake-up, voice recognition, voice calls, music playback, etc.) often only depends on the human voice part of the mixed voic...

AI Technical Summary

Problems solved by technology

The completion of many services provided by smart wearable devices (voice wake-up, voice recognition, voice calls, music playback, etc.) often only depends on the human voice part of the mixed voice, and it is the human voice of the device wearer. The presence of ambient sound, non-wearer's voice and other noise components will hinder the normal operation of some services