Speech interaction method and speech interaction equipment

A voice interaction and voice technology, applied in the direction of voice analysis, voice recognition, special data processing applications, etc., can solve problems affecting user experience, wrong human-computer interaction, and unable to remove human voice interference, so as to improve user experience and avoid human interference. The effect of computer interaction

Active Publication Date: 2018-09-07
IFLYTEK CO LTD
View PDF6 Cites 34 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] It can be seen that the current human-computer interaction instructions are obtained from the user's voice through VAD technology. Such human voice interf

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 interaction method and speech interaction equipment
  • Speech interaction method and speech interaction equipment
  • Speech interaction method and speech interaction equipment

Examples

Experimental program
Comparison scheme
Effect test

no. 1 example

[0082] see figure 2 , which is a schematic flowchart of a voice interaction method provided in this embodiment, the voice interaction method includes the following steps:

[0083] S201: Use the currently acquired user voice as the voice to be identified.

[0084] After the human-computer interaction function of the smart device is activated, the smart device will receive and recognize the user's voice in real time. For ease of distinction, in this embodiment, the currently acquired user voice is defined as the voice to be identified.

[0085] For example, if figure 1 As shown, if the currently acquired user voice is "Don't want Jacky Cheung, I want Mike to learn rock English version", then this voice is the voice to be identified; if the currently acquired user voice is "Andy Lau", then this item The voice is the voice to be identified.

[0086] S202: According to the correlation between the speech to be recognized and the historical interaction data, determine whether th...

no. 2 example

[0094] In this embodiment, the following S302 will be used to illustrate the specific implementation manner of S202 in the above-mentioned first embodiment.

[0095] see image 3 , which is a schematic flowchart of a voice interaction method provided in this embodiment, the voice interaction method includes the following steps:

[0096] S301: Use the currently acquired user voice as the voice to be identified.

[0097] It should be noted that this step S301 is the same as S201 in the above-mentioned first embodiment, and for relevant details, please refer to the first embodiment, which will not be repeated here.

[0098] S302: According to the first association relationship and the second association relationship, determine whether the speech to be identified is instruction speech.

[0099] Wherein, the first association relationship is the semantic association relationship between the speech to be recognized and the historical response result (the historical response result...

no. 3 example

[0108] This embodiment will introduce the specific implementation manner of S302 in the second embodiment above.

[0109] In this embodiment, a speech instruction recognition model may be constructed in advance, so as to use the speech instruction recognition model to judge whether the speech to be recognized is an instruction speech.

[0110] see Figure 4 A schematic flow chart of the construction method of the voice command discrimination model shown, the construction method includes the following steps:

[0111] S401: Collect various sets of human-computer interaction data belonging to the current dialogue field, wherein some or all of the human-computer interaction data includes instruction-type user sample voices and non-instruction-type user sample voices.

[0112] In the human-computer interaction scenario, for a group of human-computer dialogues (usually including one or more rounds of dialogues), the interaction is usually aimed at a specific topic. Therefore, in th...

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 application discloses a speech interaction method and speech interaction equipment. An association relationship between a to-be-distinguished speech and historical interaction data can be analyzed, and the historical interaction data include a user speech instruction before the to-be-distinguished speech and a response result for the user speech instruction; then whether the to-be-distinguished speech is an instruction-type speech can be judged according to the association relationship between the two parts; command response can be not carried out thereon when it is judged that the speechis a non-instruction-type interference speech; and thus one time of wrong human-machine interaction is avoided, and then user experience is improved.

Description

technical field [0001] The present application relates to the field of voice technology, in particular to a voice interaction method and device. Background technique [0002] With the advancement of computer and artificial intelligence technology, human-computer interaction technology is also developing continuously, especially intelligent devices such as smart homes and robots with human-computer interaction functions will have broad development prospects. [0003] Taking robots as an example, the current human-computer interaction process generally includes the following steps: [0004] First, the robot is awakened by a fixed command word; then, the robot uses the acoustic voice endpoint detection technology (Voice Activity Detection, referred to as VAD) to detect valid voice segments from the continuous voice stream, that is, to detect the starting point of valid voice. When the end point is detected, the effective speech recognition result is sent as an instruction to t...

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
IPC IPC(8): G06F17/30G10L15/22G10L15/26
CPCG10L15/22G10L15/26
Inventor 李锐陈志刚刘权王智国
Owner IFLYTEK CO LTD
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