JP-2026075028-A - Programs, devices, and methods for assisting users with mouth and/or tongue exercises.

Abstract

[Problem] To provide a program to assist users with mouth and/or tongue exercises. [Solution] When the program of the present invention is executed by the processor unit of the device, it causes the processor unit to at least receive the user's voice input, recognize at least one monosyllabic word contained in the user's voice input, and display content associated with the recognized at least one monosyllabic word on the device. In one embodiment, recognizing at least one monosyllabic word may include recognizing at least one monosyllabic word using a speech recognition model suitable for recognizing multiple different monosyllabic words spoken in succession in a short period of time and/or the same monosyllabic word spoken multiple times in succession in a short period of time. [Selection Diagram] Figure 1B

Inventors

• 十河 基文
• 西願 雅也
• 水篠 公範

Assignees

• 株式会社アイキャット

Dates

Publication Date

20260507

Application Date

20250212

Claims (1)

1. The invention described herein.

Description

An application for application of Article 30, Paragraph 2 of the Patent Law was filed. This was presented at the 15th Annual Meeting of the Japan Digital Dentistry Society, held on May 12, 2024, at Benex Nagasaki Brick Hall. This invention relates to a program, device, and method for assisting a user with oral and/or tongue exercises. The decline of the mouth area (so-called oral frailty) has been a subject of attention for some time (see, for example, Non-Patent Document 1). Japan Dental Association, “Oral Frailty,” [online], [searched September 26, 2024], Internet <https://www.jda.or.jp/enlightenment/oral/about.html> A diagram showing an example of a screen displayed on a device operated by a user.This diagram shows another example of a screen displayed on a device operated by a user.This diagram shows another example of a screen displayed on a device operated by a user.A diagram showing an example configuration of a device 300 for assisting a user with mouth and/or tongue exercises.This figure shows an example of processing performed in device 300.This figure shows another example of processing performed in device 300. The following definitions are used in this specification. A "monosyllabic word" refers to a word that has a single syllable consisting of a single vowel, or a single syllable consisting of a combination of a single consonant and a single vowel. Examples of monosyllabic words include, but are not limited to, "pa," "ta," "ka," and "ra." "Mouth and/or tongue exercises" refer to exercises that stimulate the muscles necessary for moving the mouth and/or the tongue. Therefore, "mouth and/or tongue exercises" include both exercises performed with vocalization and those performed without vocalization. The embodiments of the present invention will be described below with reference to the drawings. 1. Support for user oral and/or tongue exercises The applicant believed that improving oral frailty requires an integrated approach of oral frailty testing and training. Furthermore, considering the recent interest in games among the elderly and the expected increase in the number of elderly people who can use smartphones, the applicant believed that attempting to improve oral frailty through a game application that can be run on the devices of the elderly would be the best approach. One exercise that contributes to improving oral frailty involves repeatedly pronouncing a specific monosyllabic word multiple times in quick succession. However, conventional speech recognition models have been unable to recognize that a specific monosyllabic word has been repeatedly pronouncing it in quick succession, even when the user pronounces it multiple times in quick succession. This is because conventional speech recognition models cannot predict that the same monosyllabic word will be received multiple times in quick succession. For example, if a user pronounces the specific monosyllabic word "pa" as "papapapapa" in quick succession, a conventional speech recognition model, after receiving the initial "papa" input, cannot predict that "pa" will be pronounced again immediately after "papa," leading to a breakdown and the inability to recognize what is being input. Therefore, the applicant has developed a new speech recognition model suitable for recognizing multiple monosyllabic words pronounced consecutively in quick succession, and the same monosyllabic word pronounced multiple times in quick succession. Furthermore, the applicant found that stabilizing the volume of the user's voice input is also important in recognizing multiple monosyllabic words spoken in quick succession and the same monosyllabic word spoken multiple times in quick succession. If the volume of the user's voice input is unstable, even if the user speaks a monosyllabic word multiple times in quick succession (e.g., "pa-pa-pa-pa"), the device can only recognize that the user has spoken the monosyllabic word only once (e.g., "pa"). This is because, for example, if the volume of the user's voice input continues to exceed a predetermined threshold, a monosyllabic word spoken multiple times in quick succession (e.g., "pa-pa-pa-pa") will be recognized as a single long monosyllabic word (e.g., "paaaa"), and a single long monosyllabic word (e.g., "paaaa") will be considered a single short monosyllabic word (e.g., "pa") as a monosyllabic voice input. Therefore, stabilizing the volume of the user's voice input is necessary for the device to more accurately recognize the monosyllabic words spoken by the user. 2. Screen displayed on a user-operated device Figure 1A shows an example of a screen displayed on a user-operated device. The screen 110 shown in Figure 1A is a game screen that operates in response to the user's voice input to assist the user with mouth and/or tongue exercises. The screen 110 may be displayed on the device when a program to assist the user with mouth and/or tongue exercises is launched on the device and a game for mouth and/or tongue