Training and recognition method and device for customized product language model
A language model and training method technology, applied in the field of language model training and recognition, can solve the problems of long training time, limited, and product design that does not fully consider user needs, and achieve the effect of low coupling and high recognition rate.
Active Publication Date: 2022-04-15
AISPEECH CO LTD
View PDF0 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0004] (1) It only supports hot word update recognition, which is far from meeting the needs of developers for high customization;
[0005] (2) The training time is long, usually several hours
[0007] (1) The product design does not fully consider the needs of users, and the design of the language model architecture is insufficient;
[0008] (2) Limited by the language model training algorithm
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 moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0025] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments It is a part of embodiments of the present invention, but not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without creative efforts fall within the protection scope of the present invention.
[0026] Please refer to figure 1 , which shows a flow chart of an embodiment of the training and recognition method of the customized product language model of the present application, the training and recognition method of the customized product language model of this embodiment can be applied to the recognition of dialogues with language models o...
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
Login to View More Abstract
The present invention discloses a method and device for training and identifying a customized product language model, wherein, a method for training and identifying a customized product language model includes: in response to the release or update of a product, using training corpus to train the product, wherein , the training corpus includes general corpus, hotspot corpus, and developer-customized corpus. The language model of the product includes the first decoding, the second decoding, and the third decoding; the general corpus is input to the first decoding to convert the first decoding Train it into a general language model; input the hotspot corpus into the second decoding to train the second decoding into a solution language model; input the developer's customized corpus into the third decoding to train the third decoding into a customized product language Model. The solution provided by the method and device of the present application can train the three kinds of corpus separately so as to be more targeted to different corpus and be more friendly to developers, who only need to focus on their own customization.
Description
technical field [0001] The invention belongs to the technical field of language model training and identification, and in particular relates to a training and identification method and device for a customized product language model. Background technique [0002] In related technologies, there are products related to AIUI hot word update training on the market. By uploading hot words, such as song titles, movie titles, and language model training, the recognition rate of hot words is provided. [0003] The inventor found in the process of implementing the application that the above-mentioned solutions in the prior art mainly have the following defects: [0004] (1) It only supports hot word update recognition, which is far from meeting the needs of developers for high customization; [0005] (2) The training time is long, usually several hours. [0006] The inventor also found that the above-mentioned defects are mainly caused by the following reasons: [0007] (1) The pr...
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
Login to View More Patent Type & Authority Patents(China)
IPC IPC(8): G10L15/06G10L15/00
CPCG10L15/063G10L15/005G10L2015/0631
Inventor 徐飞陈明佳
Owner AISPEECH CO LTD