Intelligent voice follow-up visit method and system, computer equipment, storage medium and program product

Abstract

The invention discloses an intelligent voice follow-up visit method and system, computer equipment, a storage medium and a program product. The method comprises the following steps: constructing a keyword dictionary related to an answer category of a preset question; collecting voice data of answers of the follow-up object to the preset questions and converting the voice data into text data; traversing the keyword dictionary, judging whether the text data contains keywords in the keyword dictionary or not, if so, recording an answer category to which an answer belongs and continuing questioning and answering according to a first preset verbal skill script, and if not, determining an intention category of the text data through text similarity calculation; and recording an answer category corresponding to the intention category and continuing questioning and answering according to a second preset verbal skill script, wherein the first preset verbal skill script is the same as or different from the second preset verbal skill script. According to the method, the advantages of a keyword retrieval mode and a text similarity matching mode are combined, and the prior art is optimized and improved in the aspects of space resources and reaction speed.

Description

technical field [0001] The present invention relates to the technical field of intention recognition, and more specifically, to an intelligent voice follow-up method, system, computer equipment, storage medium and program product for reducing data distortion through intention recognition. Background technique [0002] In the medical field, in the past, it was necessary to manually dial the patient's phone number to collect information about the patient's physical condition index after surgery, the hospital's satisfaction score, and the physical examination notification. However, everyone's energy is limited, and because of Reasons such as personal emotions affect the patient experience. Now the intelligent follow-up system can solve this problem very well. The patient's phone information is imported into the intelligent follow-up system. Through the production of speech scripts, with the support of ASR and NLP, the machine and patient can be well realized. Dialogue, and col...

AI Technical Summary

Problems solved by technology

However, when the patient's intention is difficult to write with a few keywords, or has not appeared in the historical data, then the patient's intention at this time cannot be recognized, which will also lead to the failure to collect effective information of the patient, resulting in data distortion

[0004] There are also some intelligent follow-up systems that use text similarity matching to identify patients' intentions. However, as the historical data of patients increases, the complexity of identifying new patient intentions will increase. Some simple and clear patient intentions A few keywords can be accurately identified, but complex models are used for matching, which will waste a lot of resource space and slow down the feedback speed of the intelligent follow-up system to patients' answers, causing bad experience for patients

Images