FreeSpeech AAC is an open-source assistive communication application designed for non-verbal individuals. It leverages cloud-based AI services and large language models to provide advanced communication capabilities, serving as a cost-effective alternative to commercial AAC software.

How It Works

The application utilizes a Svelte-based frontend for a fast user interface. Backend functionalities are powered by a combination of AWS, Microsoft Cognitive Services, and Google Cloud for AI-generated voices. OpenAI's GPT-3 model handles complex textual analysis, phrase generation, and word conjugation, enabling sophisticated communication assistance.

Quick Start & Requirements

• Install dependencies: yarn install
• Run development server: yarn dev
• Requires Node.js and Yarn.
• Access to cloud service API keys (AWS, Microsoft Cognitive Services, Google Cloud, OpenAI) is necessary for full functionality.

Highlighted Details

• Integrates cutting-edge AI for voice generation and natural language processing.
• Developed with Svelte for a high-performance frontend experience.
• Positioned as a viable, lower-cost alternative to expensive commercial AAC solutions.
• Featured by CNN and BBC.

Maintenance & Community

• Project founded and led by Archer Calder.
• Core team includes Bailey Townsend.
• Community engagement is encouraged via Discord.

Licensing & Compatibility

• MIT Licensed.
• Permissive license suitable for commercial use and integration with closed-source projects.

Limitations & Caveats

The project's reliance on external, potentially costly, cloud AI services means that operational costs can be incurred by the user. Full functionality is dependent on obtaining and configuring API keys for these third-party services.