Method for providing speech and intelligent computing device controlling speech providing apparatus

Patent metadata

Field	Value
Publication number	US-11580953-B2
Application number	US-201916554374-A
Country	US
Kind code	B2
Filing date	Aug 28, 2019
Priority date	Jul 18, 2019
Publication date	Feb 14, 2023
Grant date	Feb 14, 2023

Abstract

A method for providing a speech and an intelligent computing device controlling a speech providing apparatus are disclosed. A method for providing a speech according to an embodiment of the present invention includes obtaining a message, converting the message into a speech, and determining output pattern based on a generation situation of the message, so that it is possible to more realistically convey a situation at a time of message generation to a receiver of TTS. One or more of the voice providing method, devices, intelligent computing devices controlling the voice providing device, 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. A method for providing a speech by an intelligent speech providing apparatus, the method comprising:
obtaining a message;
receiving, from a network, downlink control information (DCI) used for scheduling transmission of information related to a generation situation of the message obtained from at least one sensor included in the intelligent speech providing apparatus;
converting the message into a speech; and
providing the speech,
wherein the converting the message into a speech includes:
generating output pattern information based on the information related to the generation situation of the message, and
converting the message into a speech based on the output pattern information,
wherein the information related to the generation situation of the message is transmitted to the network based on the DCI.

2. The method of claim 1 , wherein the information related to the generation situation of the message includes information related to a creator of the message.

3. The method of claim 2 , wherein the information related to the generation situation of the message includes information related to a surrounding environment at a time the message is created.

4. The method of claim 3 , wherein the information related to the generation situation of the message includes information related to a receiver of the message.

5. The method of claim 1 , further comprising:
displaying the message on a display based on the information related to the generation situation of the message.

6. The method of claim 5 , wherein the displaying includes displaying a background image on a background of the message based on information related to a time at which the message is created or weather at a time at which the message is created.

7. The method of claim 5 , wherein the displaying includes adjusting a position of the message based on information related to an emotion of a creator at a time of creation of the message.

8. The method of claim 5 , wherein the displaying includes:
when the message is obtained using a speech signal, adjusting a distance between a plurality of syllables included in the message based on a time-domain waveform of the speech signal.

9. The method of claim 5 , further comprising:
receiving a touch input to the displayed message; and
modifying the generated output pattern information based on the touch input.

10. The method of claim 5 , further comprising:
outputting background music through an output device based on information related to a surrounding environment at a time the message is created.

11. The method of claim 1 , wherein the generating the output pattern information includes:
obtaining the output pattern information as output of a pre-learned artificial neural network by inputting the message and the information related to the generation situation of the message to the artificial neural network.

12. The method of claim 11 , wherein the artificial neural network is pre-learned by using information related to a plurality of speakers and call speech data between the plurality of speakers before the obtaining the message.

13. The method of claim 12 , wherein the generating the output pattern information further includes classifying a plurality of speeches uttered by the plurality of speakers included in the message using the artificial neural network.

14. The method of claim 1 , further comprising:
performing an initial connection procedure with the network based on a synchronization signal block (SSB), and
wherein the information related to the generation situation of the message is transmitted to the network through a physical uplink shared channel (PUSCH), and
wherein the SSB and a demodulation reference signal (DM-RS) of the PUSCH are quasi co-located(QCL) for QCL type D.

15. The method of claim 1 , further comprising:
controlling a communication unit to transmit the information related to the generation situation of the message to an artificial intelligence (AI) processor included in the network; and
controlling the communication unit to receive AI processed information from the AI processor,
wherein the AI processed information includes the output pattern information generated based on the information related to the generation situation of the message.

16. An intelligent computing device for controlling a speech providing apparatus, the intelligent computing device comprising:
a communication unit configured to obtain a message;
a processor; and
a memory including at least one command executable by the processor,
wherein the processor is configured to:
obtain information related to a generation situation of the message from the message,
receive, from a network, downlink control information (DCI) used for scheduling transmission of the information related to the generation situation of the message,
generate output pattern information based on the information related to the generation situation of the message, and
output the message to a speech based on the output pattern information,
wherein the information related to the generation situation of the message is transmitted to the network based on the DCI.

17. The intelligent computing device of claim 16 , wherein the processor applies a pre-stored user preference output pattern to the message, and
updates the output pattern applied to the message based on the information related to the generation situation of the message.

18. The intelligent computing device of claim 17 , wherein the processor obtains the information related to the generation situation of the message by inputting the message to a pre-learned first artificial neural network, and
obtains the output pattern information by inputting the information related to the generation situation of the message to a pre-learned second artificial neural network.

19. A non-transitory computer readable recording medium stored with a computer-executable component configured to execute on one or more processors of a computing device, the computer-executable component is configured to:
obtain a message;
receive, from a network, downlink control information (DCI) used for scheduling transmission of information related to a generation situation of the message obtained from at least one sensor included in an intelligent speech providing apparatus;
generate output pattern information based on the information related to the generation situation of the message;
convert the message into a speech based on the output pattern information; and
control a speech providing apparatus characterized in providing the speech,
wherein the information related to the generation situation of the message is transmitted to the network based on the DCI.

Assignees

• Lg Electronics Inc

Inventors

• Park Mihyun

Classifications

• G06N3/09

  Supervised learning · CPC title

• G06N3/0499

  Feedforward networks · CPC title

• G10L13/00 Primary

  Speech synthesis; Text to speech systems · CPC title

• G10L13/08 Primary

  Text analysis or generation of parameters for speech synthesis out of text, e.g. grapheme to phoneme translation, prosody generation or stress or intonation determination · CPC title

• G06N7/01

  Probabilistic graphical models, e.g. probabilistic networks · CPC title