Speech synthesis method, electronic device and smart home system

A speech synthesis, speech technology, applied in speech synthesis, speech analysis, audio data retrieval and other directions, can solve the problem of inability to distinguish the pronunciation of polyphonic words or false words in sentences

Active Publication Date: 2019-09-20
GREE ELECTRIC APPLIANCES INC +1
View PDF11 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The technical problem to be solved by the present invention is: in the prior art, when the speech synthesis system uses the method of probability to judge the pronunciation of polyphonic characters or Tongjiazi, when the sentence to be synthesized lacks components, it is impossible to distinguish the polyphonic characters or Tongjiazi in the sentence Pronunciation problem

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Speech synthesis method, electronic device and smart home system
  • Speech synthesis method, electronic device and smart home system
  • Speech synthesis method, electronic device and smart home system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0049] This embodiment provides a speech synthesis method, which uses a more rigorous method to determine the speech to be synthesized and perform speech synthesis. figure 1 It is an overall flowchart of the speech synthesis method of the embodiment of the present invention, as figure 1 As shown, the speech synthesis method mainly includes the following steps S101 to S106.

[0050] Before implementing the speech synthesis method, it is necessary to construct a knowledge graph in advance, and store the required voice collection through the constructed knowledge graph, figure 2 It is a flow chart of the method for constructing a knowledge map according to an embodiment of the present invention, such as figure 2 As shown, the step S101 of building a knowledge graph includes the following substeps S1011 to S1014.

[0051] In step S1011, according to multiple usage scenarios of the user, tags corresponding to each usage scenario are determined, and a connection between the usag...

Embodiment 2

[0074] This embodiment provides an electronic device, which can apply the above speech synthesis method, Figure 4 It is a block diagram of the workflow of the electronic device in the embodiment of the present invention, such as Figure 4 As shown, the electronic device mainly includes the following functional modules.

[0075] The voice recognition server 101 is configured to acquire voice demand data input by a user.

[0076] The semantic understanding server 102 is configured to analyze the voice demand data, obtain user intentions, and extract quasi-intention tags from the user intentions.

[0077] The knowledge graph server 103, wherein a knowledge graph is stored, a plurality of tags and a voice collection associated with each tag are stored in the knowledge graph, the knowledge graph server 103 is configured to determine the target intent tag according to the quasi-intent tag, from the constructed knowledge graph Query the voice collection associated with the target ...

Embodiment 3

[0095] This embodiment provides a smart home system, including one or more electronic devices mentioned above, and the user can control any electronic device as needed, making life more convenient and interesting.

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention discloses a speech synthesis method, an electronic device and a smart home system.The speech synthesis method comprises the steps that user voice requirement data areacquired; a user intention isobtained, and a quasi-intentional tag isextracted from the userintention; a text result associated with the user intention isacquired; and a target intentiontag is determined from the quasi-intentional tag, a set of speech associated with the target intentiontag isqueried from a knowledge mapping, a speech associated with each characterisdetermined based on the set of speech, and the charactersaresynthesized with the determined speech. The electronic device implements thespeech synthesis method by a speech recognition server, a semantic understanding server, a knowledge mapping server and a speech synthesis terminal, and the smart home system includes one or more of the electronic devices. Through the tags and the set of speechstored in the knowledge mapping, after each character associated with theuserintention isacquired, the speech associated with each character is determined based on the set of speech, andthecharacters and the determined speech can be synthesized into the speech with high pronunciation accuracy.

Description

technical field [0001] The present invention relates to the technical field of speech synthesis, in particular to a speech synthesis method, and also to an electronic device and a smart home system using the speech synthesis method. Background technique [0002] Modern Chinese contains many polyphonic characters, such as "tune", which can be pronounced as "diào", such as "tone", or "tiáo", such as "mediation". In the existing speech synthesis (TTS, ie text to speech) process, speech synthesis systems mostly use probabilistic methods to judge the pronunciation of polyphonic characters. [0003] However, the above-mentioned judging method has the following defects: when there is a lack of components in the sentence to be synthesized, it is impossible to distinguish the pronunciation of polyphonic characters or Tongjia characters in the sentence. For example, when "为" appears in the sentence "已为你XXX", it is difficult to distinguish whether the Chinese character "为" is pronounc...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Applications(China)
IPC IPC(8): G10L13/08G06F16/68
CPCG10L13/08G06F16/686
Inventor 吴伟贾巨涛黄姿荣秦子宁张伟伟
Owner GREE ELECTRIC APPLIANCES INC
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap