Last Updated: June 2026

1. Acceptance of Terms By downloading, installing, or using DocuVoice ("the App"), you agree to be legally bound by these Terms of Use. If you do not agree to these terms, please do not access or use the application.

2. Description of Service & Content Disclaimer DocuVoice utilizes on-device layout-aware text extraction and local neural voice synthesis (ONNX) to convert user-supplied PDF documents into spoken audio with interactive visual sentence highlighting.

• Personal Productivity and Convenience: The application is designed strictly for personal productivity, study, accessibility, and reading convenience.
• Copyright Responsibility: DocuVoice is a tool that allows users to listen to documents they own or have the legal right to access. DocuVoice does not host, index, share, or distribute any third-party documents. You are solely responsible for ensuring that you possess the necessary legal rights, copyrights, or licenses to import, process, and listen to the PDF files you load into the App.
• Local Neural Engine Limitations: Speech synthesis is completed entirely locally on-device. While we strive for high accuracy and natural-sounding pronunciation, we do not guarantee perfect transcription, formatting, pronunciation, or translation of all PDF layouts, complex mathematical formulas, code blocks, or scanned document layers.

3. Data Storage, User Content, and Conduct We prioritize your data privacy through a decentralized, offline-first architecture:

• Local Storage: All of your imported PDF files, metadata, reading logs, and progress history are stored exclusively locally on your device within an encrypted Core Data container. We do not maintain servers to store your files. If you delete the App, all your imported documents and history will be permanently lost unless backed up via your own device's iCloud system settings.
• No Cloud Processing: All document text extraction, layout reconstruction, and voice generation occur locally on your device. No document text or speech data is sent to external cloud servers.
• User Conduct: You agree to only import files that you have the legal right to process. You agree not to reverse-engineer, decompile, or attempt to extract the local voice models, ONNX runtime structures, or proprietary layout-aware parsing algorithms.

4. Subscriptions and In-App Purchases DocuVoice is currently fully unlocked. All premium features, local high-quality AI voices, and unlimited document processing are available free of charge. If in-app purchases or subscriptions are introduced in the future, billing will be handled exclusively by Apple under standard iOS App Store account terms.

5. Intellectual Property All original app designs, UI/UX elements, reader interfaces, branding, custom layout-aware parsing algorithms, source code, and local ONNX voice configurations are the intellectual property of the DocuVoice development team. Unauthorized copying, reverse-engineering, or distribution is strictly prohibited.

6. Limitation of Liability To the maximum extent permitted by applicable law, DocuVoice and its developers shall not be liable for any direct, indirect, or consequential damages (including, but not limited to, data loss, document corruption, copyright disputes, hardware performance degradation, or emotional/intellectual outcomes) arising out of the use or inability to use the application. Use of the App and reliance on its text extraction and voice synthesis is entirely at your own risk.

7. Third-Party Frameworks and AI Models The App utilizes local third-party runtimes (such as Microsoft OnnxRuntime) to generate voice waveforms. We are not responsible for software crashes, model performance bugs, or inaccuracies in voice synthesis arising from these underlying libraries.

8. Governing Law These terms shall be governed by and construed in accordance with the laws of the jurisdiction in which the developer operates, without regard to its conflict of law provisions, and adhering to standard international mobile software regulations.