People’s memory decreases as the years go by. In daily life, elderly people often get trapped in many troubles because they "can't remember", such as forgetting where their identity documents are, forgetting whether they have taken their medication and when they should take it. This paper analyzes the background of the problem and the current situation of the research, defines the system functions needed to address the problem, and designs and implements a voice-activated APP, Item Finder, for the elderly to record storage information of items.
In order to improve the recognition rate of elderly speech and minimize the processing memory occupation and time consumption, the system introduces the Kalman filter, a data processing algorithm commonly used in control and navigation fields, and proposes a human speech enhancement algorithm. In addition, the system builds a speech recognition grammar through BNF paradigm and adopts the speech technology from iFLYTEK platform to realize speech recognition processing. Meanwhile, this paper fully adheres to the goal of aging appropriateness, analyzes the physiological characteristics of the elderly user group, focuses on designing the product aging appropriateness strategy, and tailors the interaction scheme for elderly users.
The system is developed using the Android platform and is hosted on a cloud server using the FastAPI web framework. It is also improved with supplementary functions such as reminders, security verification, personalized interface, login, and registration to enhance the integrity of the system. The research work in this paper can effectively help the elderly to solve the problem of item storage management, which has great application value and market prospect.
Key words: Aging-friendly; Item management; Android; Speech recognition
Next time recall the latest origin commit
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent