Voice query method and device, storage medium and terminal equipment
A technology of voice and query object, applied in the field of computer
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0074] see figure 1 , the embodiment of the present invention provides a voice query method, including step S10 to step S50, specifically as follows:
[0075] S10. Receive a user's voice query request.
[0076] In the embodiment of the present invention, it can be executed by the user terminal, which directly receives the user's voice and performs voice interaction with the user; it can also be executed by the server, then the user can be referred to as the user terminal, and the user terminal receives the user's voice and transmits the voice signal Send it to the server for voice recognition and perform subsequent query operations. The user terminal includes, but is not limited to, a mobile phone, a wearable device, a vehicle-mounted device, and the like.
[0077] S20, perform voice recognition on the voice query request, and obtain the query sentence.
[0078] In the embodiment of the present invention, the voice query request is a voice signal, which can be converted by ...
Embodiment 2
[0094] see image 3 , On the basis of Embodiment 1, this embodiment provides an intent classification model and an object recognition model for intent recognition and object recognition, that is, the specific implementation process of step S30 in Embodiment 1 includes:
[0095] S31. Identify the query statement according to the intent classification model, and obtain the query intent of the query statement.
[0096] In this embodiment, the intent classification model can be trained by machine learning, that is, query sentences and intent words are associated to form training data pairs, and then a large amount of training data pairs are input into the intent classification model for training.
[0097] S32. Identify the query statement according to the query intent and the object recognition model, and obtain query object words of the query statement.
[0098] Since the query statement contains multiple words, when the object recognition model is used to identify the slot of t...
Embodiment 3
[0113] see Figure 8 , the embodiment of the present invention also provides a device for voice inquiry, including:
[0114] Voice query receiving module 10, for receiving the user's voice query request;
[0115] Query statement obtaining module 20, is used for carrying out speech recognition to described voice query request, obtains query statement;
[0116] The intent and object recognition module 30 is used to perform intent recognition and object recognition on the query sentence, and obtain the query intent and query object words of the query sentence; wherein, the query object words correspond to at least one query object;
[0117] A query object confirmation module 40, configured to determine the query object of the query statement based on the query object words; and
[0118] The query database module 50 is configured to query a retrieval database according to the query intention and the query object of the query statement, and obtain a query result.
[0119] With r...
PUM
Abstract
Description
Claims
Application Information
- R&D Engineer
- R&D Manager
- IP Professional
- Industry Leading Data Capabilities
- Powerful AI technology
- Patent DNA Extraction
Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.
© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com