Skip to content

About

Archived SwiftUI research prototype for prepared AAC phrases, Apple Personal Voice and wearable audio (ASSETS 2024).

Topics

Resources

Stars

0 stars

Watchers

1 watching

Forks

Latest commit

 

History

3 Commits

Folders and files

Repository files navigation

Prompt AAC

Research archive: this prototype was created for a 2024 study and is no longer actively maintained. It is preserved to document the design and support research reuse; it is not a production AAC system or a clinical device.

Prompt AAC interface and wearable speaker overview

Prompt AAC is a SwiftUI prototype for preparing short phrases and playing them through a small wearable Bluetooth speaker. A message can act as a private prompt for the wearer or as public speech during a communication breakdown, while the movable speaker brings audio closer to a conversation partner.

The app supports standard system voices and Apple Personal Voice, allowing prepared phrases to use a synthesised voice that sounds like the speaker. It was co-designed and evaluated with people with aphasia as part of the Screenless AAC research programme at King's College London.

What the prototype includes

  • Creation and deletion of reusable text prompts
  • Local JSON persistence on the device
  • Speech playback through AVSpeechSynthesizer
  • Standard voice selection and experimental Personal Voice access
  • System audio routing to a connected Bluetooth speaker
  • SwiftUI source for iPhone and iPad

Bluetooth pairing and routing are handled by iOS; this repository does not contain a custom Bluetooth protocol. The included watchOS target is an unused Xcode template and was not part of the evaluated prototype.

Run locally

  1. Clone the repository.
  2. Open PromptAAC.xcodeproj in Xcode.
  3. Select the PromptAAC scheme and an iPhone or iPad running iOS/iPadOS 17.4 or later.
  4. Configure your own signing team if you want to run on a physical device.
  5. Pair a Bluetooth speaker in the device's system settings if you want external audio.

The project has no third-party package dependencies.

Personal Voice

Personal Voice must first be created in the device's accessibility settings. The user must then allow the app to request access. Availability depends on Apple's supported hardware, operating systems and languages; see Apple's current setup guidance.

Repository map

PromptAAC/
├── Data/                 Codable prompt model
├── ContentView.swift     Voice selection and entry point
├── AudioPromptView.swift Saved prompts and speech playback
└── EditAudioView.swift   Prompt creation and persistence

Research

If you use the prototype in academic work, please cite the accompanying paper; structured citation metadata is available in CITATION.cff.

Limitations

This is a study prototype rather than a deployable AAC product. Prompts are stored only on the device, Personal Voice access is experimental, and the app does not include cloud sync, clinical configuration or production-grade error handling. No participant data is included in the repository.

Licence

The source code is available under the MIT Licence.

About

Archived SwiftUI research prototype for prepared AAC phrases, Apple Personal Voice and wearable audio (ASSETS 2024).

Topics

Resources

Stars

0 stars

Watchers

1 watching

Forks

Releases

Packages

Contributors

Languages