AI Policy

No LLM-powered assistants, such as ChatGPT, Gemini, etc., are used for any purpose to create TypingTalk AAC. Every line of code in both the Android and iOS versions is written, debugged, and maintained by a human (me). This is purely to help me closely oversee the app's contents and functionality.

The goal of my app's integration with text-to-speech is to provide ways for users with speech difficulties (like me) to communicate and express themselves, while avoiding wasteful energy usage and malicious applications such as impersonation.

The Google Text-to-Speech voices in both versions are legacy models providing low-cost and quick text-to-speech, especially useful for real-time communication, while still sounding natural and based on the recordings of consenting voice actors.

Apple Personal Voice, which can be used with my iOS app, allows users to create a text-to-speech model based on their own voice, which is stored and runs completely offline on the user's device, and provides security measures to deter malicious use.

Feel free to contact me if you have any other questions. Thank you so much.