Beyond Words
Privacy Policy
How Beyond Words processes account information, user-provided text, service usage, diagnostics, and subscription data.
Effective date: 2026-07-29
Last updated: 2026-09-24
This Privacy Policy explains how Beyond Words processes information when you use the Android application and related services. The service is provided by Yu Tang, operating as Independent Developer in China.
1. Scope
This policy applies to the Beyond Words mobile application, its cloud-backed
features, its support channels, and the application pages hosted under
apps.tangyu.me.
It does not govern third-party services that you choose to access independently, including websites or applications opened from external links.
2. Information processed
2.1 Account information
You do not create a user-facing account or sign in with an email address, password, or social-login provider. To provide cloud-backed features, the application automatically creates and processes a Firebase anonymous user ID. We use this pseudonymous identifier for authentication, quota enforcement, fraud and abuse prevention, and purchase-entitlement verification. If user-facing account features are introduced, this policy will be updated before those features are enabled.
2.2 Text and generated content
When you request a transformation, the application processes:
- the text you submit;
- the selected context pack or transformation mode;
- configuration needed to produce the result;
- the generated output;
- technical request metadata such as timestamps, request identifiers, and success or error status.
For the built-in model service, submitted text, the selected context-pack instructions, and the information needed to generate a response are sent through our Firebase backend to Google Gemini. Our backend does not retain ordinary transformation text or generated output as long-term content records.
If you choose an OpenAI-compatible custom model service, the application sends the submitted text, selected context-pack instructions, and generation request directly from your device to the HTTPS endpoint that you configure. Your API key is sent to that provider only to authenticate the request; it is not sent to our Firebase backend. That provider’s handling, retention, locations, and charges are governed by its own terms and privacy practices.
When you use voice input, the recording is sent to our Firebase backend and to Google Gemini to create a transcription. Our backend does not persist the audio recording, transcription text, or provider response after the transcription operation. Google Gemini processes the recording under its applicable terms and privacy practices.
Do not submit passwords, payment card details, government identifiers, confidential business information, health records, or other information that you are not authorized to process.
2.3 Local history
Transformation history is stored locally on your device so you can reopen and refine previous results. It is not synchronized to another device unless a future synchronization feature is explicitly introduced and described in an updated policy.
Imported custom context packs and custom-model configuration are also stored on your device. A custom-model API key is stored using platform secure storage and is not stored in ordinary application preferences.
2.4 Usage, quota, and entitlement information
We process information needed to enforce usage limits and subscription entitlements, including the anonymous user ID, quota balance, renewal or reset period, request counts, subscription product identifiers, purchase state, and verification results.
Payment card information is handled by Google Play and is not provided directly to Independent Developer.
2.5 Diagnostics and analytics
Automated crash-reporting collection is disabled in the current configuration. Support requests may still include technical information that you choose to provide.
General behavioral analytics are disabled in the current configuration.
The application does not display third-party advertising and does not process data for targeted advertising.
2.6 Support communications and content reports
When you contact support, we process the contact details, message, attachments, and technical information that you choose to provide so that we can investigate and respond.
If you submit a harmful-output report, we process and store the generated output, the report category, an optional note, and limited generation context. You may also choose to include the original input text with that report. These records are used for safety review, investigation, and improvement; they are not created for ordinary transformations.
3. How information is used
Information is processed to:
- authenticate anonymous service sessions and protect access;
- perform requested text transformations;
- calculate quotas and verify subscription access;
- provide local history and user-requested restoration;
- prevent abuse, fraud, and unauthorized access;
- diagnose service failures and improve reliability;
- respond to support, privacy, and harmful-output reports;
- comply with legal obligations and enforce the Terms of Service.
We do not sell personal information.
4. Service providers
The service relies on providers that process information on our behalf or under their own terms, including:
- Google Firebase — Firebase Authentication, Cloud Functions, Firestore, and App Check; Firebase Authentication provides anonymous authentication, App Check helps protect the service, Cloud Functions process cloud-backed requests, and Firestore stores the records described in this policy;
- Google Play — application distribution, subscription billing, purchase records, and entitlement information;
- Google Gemini — generation for the built-in model service and voice transcription. See the Gemini API Additional Terms of Service and the Google Privacy Policy for Google’s applicable data-handling terms;
- a custom model service provider that you select — direct processing of generation requests when you choose to configure and use that service;
- infrastructure or email providers used to host the public pages and respond to support requests.
These providers may process information in countries other than your own. Their contractual terms and technical controls govern their processing.
5. Retention
We retain information only for as long as needed for the purposes described in this policy.
- Local history remains on your device until you delete it, clear application data, or uninstall the application.
- Imported context packs, custom-model configuration, and its locally stored API key remain on your device until you remove them, clear application data, or uninstall the application.
- Completed usage events and quota ledgers are retained for up to 90 days after the relevant event or quota period. They do not contain ordinary transformation text or generated output.
- Harmful-output reports, including the content that you choose to submit with them, are retained for 90 days.
- Purchase and entitlement records are retained while needed to provide and verify an active entitlement. After an entitlement ends, related records are generally retained for up to 24 months for fraud prevention, dispute resolution, accounting, and legal compliance. An active lifetime entitlement is retained while it remains active so that it can be verified or revoked.
- Server and security logs are retained only for the period reasonably needed for security, fraud prevention, and service operation.
- AI-provider handling and retention are subject to the provider’s applicable terms and privacy practices.
6. Sharing and disclosure
Information may be disclosed:
- to the providers listed above as necessary to operate the service;
- when you direct us to share or export content;
- to investigate security incidents, abuse, or violations of the Terms;
- when required by law, legal process, or a valid governmental request;
- as part of a merger, acquisition, restructuring, or transfer of the service, subject to appropriate notice and safeguards.
7. Security
We use reasonable administrative, technical, and organizational measures, including authenticated backend access, restricted service credentials, transport encryption, and HTTPS-only custom-model endpoints. Custom-model API keys are stored on the device using platform secure storage. No system is completely secure, and we cannot guarantee that unauthorized access will never occur.
8. Your controls and rights
Depending on your location, you may have rights to access, correct, delete, restrict, or object to certain processing, and to obtain a copy of personal information.
You can:
- edit or delete locally stored history in the application;
- manage or cancel subscriptions through Google Play;
- contact us at support@tangyu.me for a privacy request.
Because the application does not provide a user-facing account, we may need information that helps us locate relevant server-side records before completing a request. We will respond to privacy requests in accordance with applicable law.
9. Changes to this policy
We may update this policy when the service, providers, or legal requirements change. The revised policy will identify a new “Last updated” date. Material changes may also be communicated in the application or through another appropriate channel.
10. Contact
Yu Tang , China
Support: support@tangyu.me