Intelligent voice recognizing method, apparatus, and intelligent computing device

Patent metadata

Field	Value
Publication number	US-11114093-B2
Application number	US-201916555659-A
Country	US
Kind code	B2
Filing date	Aug 29, 2019
Priority date	Aug 12, 2019
Publication date	Sep 7, 2021
Grant date	Sep 7, 2021

Abstract

An intelligent voice recognition method, voice recognition apparatus and intelligent computing device are disclosed. An intelligent voice recognition method according to an embodiment of the present invention obtains a microphone detection signal, recognizes a voice of a user from the microphone detection signal and outputs a response related to the voice on the basis of a result of recognition of the voice, wherein the microphone detection signal includes noise, and a microphone detection signal including only the voice obtained by removing the noise from the microphone detection signal is recognized. Accordingly, only a voice of a user can be effectively separated from a microphone detection signal detected through a microphone of the voice recognition apparatus. One or more of the voice recognition devices, intelligent computing devices, and servers of the present invention may include artificial intelligence modules, drones (Unmanned Aerial Vehicles, UAVs), robots, Augmented Reality (AR) devices, and virtual reality (VR) devices, devices related to 5G services, and the like.

First claim

What is claimed is:

1. An intelligent voice recognition method of a voice recognition apparatus, comprising:
obtaining a microphone detection signal;
recognizing a voice of a user from the microphone detection signal; and
outputting a response related to the voice based on a result of recognition of the voice,
wherein the microphone detection signal includes noise, and the recognizing of the voice comprises recognizing a microphone detection signal including only the voice obtained by removing the noise from the microphone detection signal, and
wherein the obtaining of the microphone detection signal is obtained from a wireless earphone for detecting external signals and comprises:
receiving an uplink (UL) grant from a network;
requesting the microphone detection signal from the wireless earphone using frequency hopping set based on a number of repetitions of transmission included in the UL grant; and
obtaining the microphone detection signal in response to the requesting the microphone detection signal.

2. The intelligent voice recognition method of claim 1 ,
wherein the recognizing of the voice comprises removing ambient noise of the wireless earphone from the microphone detection signal.

3. The intelligent voice recognition method of claim 1 , wherein the outputting of the response comprises transmitting a microphone detection signal including only the voice obtained by removing the noise from the microphone detection signal to an external device connected to the voice recognition apparatus for a call.

4. The intelligent voice recognition method of claim 1 , wherein the outputting of the response comprises:
receiving resource allocation information for call connection with an external device from a network through a physical uplink control channel; and
transmitting the microphone detection signal including only the voice from which the noise has been removed to the external device based on the resource allocation information.

5. The intelligent voice recognition method of claim 1 , wherein the voice recognition apparatus includes a virtual reality device or an augmented reality device,
wherein the recognizing of the voice comprises removing ambient noise of a head module of the virtual reality device or the augmented reality device from the microphone detection signal.

6. The intelligent voice recognition method of claim 1 , wherein the recognizing of the voice comprises:
inputting the microphone detection signal to an artificial neural network learned in advance; and
obtaining the microphone detection signal including only the voice from which the noise has been removed as an output of the artificial neural network,
wherein the artificial neural network is learned in advance based on a training set including prestored sound signals and the voice of the user.

7. The intelligent voice recognition method of claim 6 , further comprising:
receiving downlink control information (DCI) used to schedule transmission of the microphone detection signal obtained from at least one sensor included in the voice recognition apparatus from a network; and
transmitting the microphone detection signal to the network based on the DCI.

8. The intelligent voice recognition method of claim 7 , further comprising:
performing an initial access procedure with the network based on a synchronization signal block (SSB); and
transmitting the microphone detection signal to the network through a PUSCH,
wherein the SSB and a DM-RS of the PUSCH are QCLed for QCL type D.

9. The intelligent voice recognition method of claim 8 , further comprising:
controlling a communication module to transmit the microphone detection signal to an AI processor included in the network; and
controlling the communication module to receive AT-processed information from the AI processor,
wherein the AI-processed information is a microphone detection signal including only the voice from which the noise has been removed.

10. An intelligent voice recognition apparatus comprising:
a microphone for detecting external signals;
a speaker for outputting sound to the outside; and
a processor configured to obtain a microphone detection signal through the microphone, to recognize a voice of a user from the microphone detection signal, and to output a response related to the voice through the speaker based on a result of recognition of the voice,
wherein the microphone detection signal includes noise, and the processor recognizes a microphone detection signal including only the voice obtained by removing the noise from the microphone detection signal, and
wherein the processor obtains the microphone detection signal from a wireless earphone for detecting external signals by:
receiving a uplink (UL) grant from a network;
requesting the microphone detection signal from the wireless earphone using frequency hopping set based on a number of repetitions of transmission included in the UL grant; and
obtaining the microphone detection signal in response to the requesting the microphone detection signal.

11. The intelligent voice recognition apparatus of claim 10 , wherein the processor removes ambient noise of the wireless earphone from the microphone detection signal.

12. The intelligent voice recognition apparatus of claim 11 , wherein the voice recognition apparatus is a first mobile terminal including a communication module, and the processor transmits a microphone detection signal including only the voice obtained by removing the noise from the microphone detection signal to a second mobile terminal connected to the first mobile terminal for a call through the communication module.

13. The intelligent voice recognition apparatus of claim 10 , wherein the voice recognition apparatus is a virtual reality device or an augmented reality device, and the processor removes ambient noise of a head module of the virtual reality device or the augmented reality device from the microphone detection signal.

14. The intelligent voice recognition apparatus of claim 10 , wherein the processor inputs the microphone detection signal to an artificial neural network learned in advance and obtains the microphone detection signal including only the voice from which the noise has been removed as an output of the artificial neural network,
wherein the artificial neural network is learned in advance based on a training set including prestored sound signals and the voice of the user.

15. The intelligent voice recognition apparatus of claim 14 , wherein the artificial neural network is learned through a supervised learning method by setting the prestored sound signals as input values and setting the voice of the user as an output value.

16. The intelligent voice recognition apparatus of claim 15 , wherein the processor separates the microphone detection signal into the voice of the user and the noise in the time domain and the frequency domain, extracts feature vectors from the voice of the user and the noise and inputs the feature vectors to the artificial neural network.

17. The intelligent voice recognition apparatus of claim 16 , wherein the processor receives downlink control information MCI) used to schedule transmission of the microphone detection signal obtained from the microphone included in the voice recognition apparatus from a network through the communication module and transmits the microphone detection signal to the network based on the DCI.

18. The intelligent voice recognition apparatus of claim 17 , wherein the processor performs an initial access procedure with the network based on a syn

Assignees

• Lg Electronics Inc

Inventors

• Kim Wonchul

Classifications

• G06N3/08

  Learning methods · CPC title

• G10L15/22 Primary

  Procedures used during a speech recognition process, e.g. man-machine dialogue · CPC title

• G10L15/063

  Training · CPC title

• H04L5/0044

  Allocation of payload; Allocation of data channels, e.g. PDSCH or PUSCH · CPC title

• G10L21/0272 Primary

  Voice signal separating · CPC title