Radio Me: Real-time Radio Remixing for people with mild to moderate dementia who live alone, incorporating Agitation Reduction, and Reminders

Radio Me：为患有轻度至中度痴呆症的独居患者提供实时广播混音，包括减少焦虑和提醒

• 批准号: EP/S026991/1
• 负责人: Eduardo Miranda
• 金额: $ 61.29万
• 依托单位: Plymouth University
• 依托单位国家: 英国
• 项目类别: Research Grant
• 财政年份: 2019
• 资助国家: 英国
• 起止时间: 2019 至 无数据
• 项目状态: 未结题
• 来源: https://gtr.ukri.org/projects?ref=EP%2FS026991%2F1
• 关键词: Radio; Me; Real; time; Remixing

The Radio Me project will deliver an aid for people with mild to moderate dementia (henceforth referred to as PWD) living alone in their homes, to provide memory reminders and agitation reduction. Radio listening is most common in the age range of PWD, and mainstream radio has two parts: played music, and spoken voice by DJs. Thus, it provides an interface involving voice and music, and one that is familiar to PWD. Radio Me (RM) is a system that maps these natural voice and music elements onto integrated aids for memory and agitation. RM will enable broadcast-style radio as a mode of human-computer interaction to support PWD who live alone in their own homes. The title Radio Me refers to the system rather than the audio resulting from the real-time generative remixing.An example of functionality is envisaged as follows: RM audio output through the speakers will, by default, sound like the live local radio. So, when the PWD switches on the radio in the morning, it initially sounds like their local station. However, at some pre-decided point (as entered into an electronic diary by PWD or their carer), and at the start of a song, a DJ-like voice seamlessly overrides the real DJ and reminds the listener to hydrate. A little later, the radio reminds the listener to eat lunch. Soon RM detects that the listener is becoming agitated (via a worn wrist-sensor). It overrides the next DJ song choice and selects a song from the user's personal library, which is known is likely to calm them. It can keep playing calming material until it detects the user has calmed. RM can give more frequent date / time checks for the listener than a normal radio station, and remind them to take their medication, to attend a Memory Café, etc.The project is timely because: (1) it addresses dementia, one of the UK's major national health and care priorities; (2) it addresses the UK's care profession crisis.

Radio Me 项目将为独居在家中的轻度至中度痴呆症患者（以下简称 PWD）提供帮助，以提供记忆提醒和减少躁动。广播收听在残疾人这个年龄段最为常见，主流广播有两部分：播放的音乐和DJ的语音。因此，它提供了一种涉及语音和音乐的界面，并且是残疾人熟悉的界面。 Radio Me (RM) 是一个将这些自然声音和音乐元素映射到集成辅助记忆和兴奋的系统。 RM 将使广播式广播作为一种人机交互方式来支持独居在家中的残疾人。标题 Radio Me 指的是系统，而不是实时生成混音产生的音频。功能示例设想如下：默认情况下，通过扬声器输出的 RM 音频听起来像现场直播的本地广播。因此，当残疾人早上打开收音机时，它最初听起来就像是他们当地的电台。然而，在某个预先确定的时间点（由残疾人或其护理人员输入电子日记），在歌曲开始时，类似 DJ 的声音会无缝地覆盖真正的 DJ 声音，并提醒听众补充水分。过了一会儿，广播提醒听众吃午饭。很快 RM 检测到听众变得焦躁（通过佩戴的手腕传感器）。它会覆盖下一首 DJ 歌曲选择，并从用户的个人库中选择一首歌曲，已知这可能会让他们平静下来。它可以继续播放平静的材料，直到检测到用户已经平静下来。 RM 可以比普通广播电台更频繁地为听众检查日期/时间，并提醒他们吃药、去记忆咖啡馆等。该项目很及时，因为：（1）它解决了痴呆症问题，这是英国主要的国家健康和护理优先事项之一； (2)它解决了英国的护理行业危机。

项目成果

Context-Aware Intelligent Mixing Systems

情境感知智能混音系统

• DOI: 10.17743/jaes.2020.0043
• 发表时间: 2021
• 期刊: Journal of the Audio Engineering Society
• 影响因子: 1.4
• 作者: Lefford M

Artificially Synthesising Data for Audio Classification and Segmentation to Improve Speech and Music Detection in Radio Broadcast

• DOI: 10.1109/icassp39728.2021.9413597
• 发表时间: 2021-02
• 期刊: ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
• 作者: S. Venkatesh;D. Moffat;Alexis Kirke;Gözel Shakeri;S. Brewster;J. Fachner;Helen Odell-Miller;Alexander J. Street;Nicolas Farina;Sube Banerjee;E. Miranda

Word Embeddings for Automatic Equalization in Audio Mixing

用于音频混合中自动均衡的词嵌入

• DOI: 10.17743/jaes.2022.0047
• 发表时间: 2022
• 期刊: Journal of the Audio Engineering Society
• 影响因子: 1.4
• 作者: Venkatesh S

Generative Music with Partitioned Quantum Cellular Automata

使用分区量子细胞自动机生成音乐

• DOI: 10.3390/app13042401
• 发表时间: 2023
• 期刊: Applied Sciences
• 作者: Miranda E

A Deep Learning Approach to Intelligent Drum Mixing With the Wave-U-Net

使用 Wave-U-Net 进行智能鼓混音的深度学习方法

• DOI: 10.17743/jaes.2020.0031
• 发表时间: 2021
• 期刊: Journal of the Audio Engineering Society
• 影响因子: 1.4
• 作者: Martínez Ramírez M