Image to Doc & Sheet AI
Privacy Policy • Last updated: 26 October 2026
Operator: Sunlight Rose Studio
Contact: nhat.ngtan@gmail.com
Android package: com.image2docsheet.ai
Delete conversion results without deleting your account
On Android or the website, sign in to the account used for the conversion and open its available Word or Excel results (https://hocveai.com/word or https://hocveai.com/excel). For a single result, select “Delete result” (“Xóa kết quả”) and confirm. For a batch, tap or click the relevant file row to expand its actions, select “Delete result” and confirm. Repeat for any other results you want to delete; deleting one result does not delete the others.
This action deletes the selected result files from our server and removes their download access without deleting your account or remaining pages. Account data, job metadata, credit ledger and transaction records are retained separately as described in section 4. It does not remove provider records or copies already downloaded or shared. Result download access expires one hour after completion. That download deadline is not a guarantee that all data is physically erased at that instant: scheduled cleanup can be delayed by outages or a backlog, and other records have separate retention rules. If deletion reports an error, do not assume it completed; retry when the service is available. For full account deletion, follow section 5.
1. About this policy
Image to Doc & Sheet AI converts selected document photos and scanned PDFs into editable Word or Excel files. This policy covers our Android app and website. Google Sites hosts this public policy page and operates under Google's own privacy terms.
2. Information we process
Account information: You can sign in with Google or email/password through Firebase Authentication. Passwords for email registration must contain at least six characters. Email verification is not required for ordinary document conversion. Firebase manages credentials, verification status and password-reset emails, and processes authentication tokens, IP addresses and user-agent/device information for authentication and security. Our conversion backend does not receive or store your password. It validates your Firebase ID token and associates its UID with your page balance and activity. Google sign-in may also supply a display name and available profile information; a display name is not required for email/password accounts. Each UID has its own balance and history. The app does not provide provider linking, automatic account merging or page transfers between UIDs.
Free-trial eligibility: While the program is enabled and capacity remains, eligible Firebase accounts created after activation may receive two free conversion pages once. Google sign-in accounts can qualify immediately; email/password accounts must verify their email to receive free pages. Verification is not required to use purchased pages or for ordinary sign-in. The initial program is limited to 100 accounts and 200 pages in total. Existing QA/reviewer accounts and accounts that have already received the offer are excluded. We use Firebase UID, account-creation and email-verification metadata, and an atomic ledger/counter to determine eligibility, prevent duplicate grants and enforce these limits. Granted pages do not have a separate expiry. Disabling new grants does not remove pages already granted or purchased. The minimal replay-prevention and ledger records follow the retention described in section 4.
Documents: We receive the photos/PDFs you choose, their file names and page counts. We process page images, OCR text, layout data and generated DOCX/XLSX files to perform your conversion. Documents may contain personal or sensitive information; upload only material you are entitled to process.
Transactions: Google Play handles payment details. Our backend processes product IDs, purchase tokens, order identifiers, purchase status and credit adjustments to verify purchases, reconcile cancellations/refunds and prevent duplicate credits. We do not receive your payment card or bank account details. On Android, page packages are available as one-time purchases through Google Play in supported markets. They are not subscriptions; unused purchased pages do not automatically expire. The app displays prices and currencies returned by Google Play. Pages are credited only after our backend verifies a valid PURCHASED transaction with Google. Pending transactions are not credited until purchased; purchase consumption permits buying a package again. Google Play license-test transactions are identified separately as TEST purchases in our records. Website checkout and Fast remain unavailable.
Operational data: We retain job identifiers, document metadata, conversion status, errors, page counts, processing time, AI token/cost usage and credit history for service operation, troubleshooting and fraud prevention. Hosting providers may process IP addresses and request/security logs. We do not intentionally write document contents or authentication/purchase tokens to application logs. We do not integrate advertising, Firebase Analytics or Crashlytics SDKs in this build.
3. How information is used and processed
We use this information to sign you in, convert your selected documents, provide downloads, maintain shared account credits, secure the service and resolve support requests. Google Cloud/Firebase provide authentication, private storage, processing, job dispatch and database services. Netlify hosts the website. Selected page images and useful extracted OCR/layout information are sent from our backend to OpenAI through its API for document reconstruction. We do not sell documents or use them for advertising. The app has no public document feed or user-to-user content hosting. If you choose to share a downloaded file, the recipient/app you select receives that copy.
OpenAI requests disable response storage (store=false). This is not a zero-retention guarantee: OpenAI's default abuse-monitoring logs may contain input/output content and be retained for up to 30 days, with longer retention possible for legal or safety reasons. Provider caching and other service retention rules may also apply. No special Zero Data Retention arrangement has been verified for this service. See https://developers.openai.com/api/docs/guides/your-data.
4. Storage, expiry and deletion
Our conversion backend/storage is deployed in Singapore. Authentication and other providers may process data in other countries, including the United States; we do not promise that all data stays in Singapore.
Server source files and processing checkpoints are deleted when processing finishes, including failed conversions. Unconfirmed uploads expire after 15 minutes. Started jobs have a recovery/expiry window of up to four hours. Result downloads expire one hour after completion. A scheduled maintenance task runs every minute to remove expired files and repair interrupted cleanup; outages or a backlog can delay physical deletion. You can delete completed results earlier using “Delete result”. These limits apply to our temporary document storage, not provider records, account/transaction data or copies you download/share.
Job metadata, credit ledger, pseudonymous account/replay-prevention records and transaction records (including server-side purchase tokens/order IDs where present) are retained separately to operate and reconcile the service and prevent fraud. They do not have the one-hour document expiry or an automatic fixed deletion deadline in the current system. Account deletion removes document/job metadata as described in section 5. After account deletion, retained records can include a hashed account identifier, account/deletion timestamps and status, purchase counts, ledger amounts and transaction references, and purchase verification/consumption/refund status and identifiers. These pseudonymous records are not claimed to be anonymous. Short-lived rate-limit counters follow their existing expiry rather than the document-deletion process. Minimal records may remain while the service operates and where required to resolve disputes or legal obligations. Contact us about access, correction or deletion requests.
On Android, capture drafts are stored privately per account/tool to recover your session. Drafts are removed when used, emptied or signed out; drafts older than one hour are cleaned when the app is reopened. Capture/cache files and downloaded/shared copies may remain on your device until removed by the app, operating system or you. Camera photos are not automatically saved to the photo gallery.
5. Account deletion and your choices
Use “Delete account” under Account → Privacy & data in the app, or visit https://hocveai.com/delete-account and sign in with the same app account using its existing Google or email/password method. Wait for active conversions to finish first. Once accepted, the account is blocked from further use and remaining pages are cleared. Deletion removes associated server document files and document/job metadata in bounded batches, then removes the Firebase sign-in account. If interrupted or a temporary error occurs, the deletion request remains pending and the existing maintenance worker retries from a stored checkpoint. A pending confirmation means the request has been accepted, not that every deletion step has completed. Completion is recorded only after the final deletion steps succeed. Minimal pseudonymous account/replay-prevention and ledger/transaction records may remain as described above. Deleting this app account does not delete your Google account or Google's independent payment records. Provider backups/security records follow their own deletion schedules.
You can also request account/data deletion by emailing nhat.ngtan@gmail.com from your registered email with the subject “Image to Doc & Sheet AI – Account deletion”. Do not send passwords, purchase tokens or document contents. We may ask for proportionate ownership verification. We do not promise an unsupported fixed completion deadline; we will explain any records that must be retained.
You choose which files to upload and when to use the camera. Camera permission is requested for capture. The app does not request location, microphone, contacts or broad photo-library permissions. You may decline camera access and select a file instead. OS/identity SDK permissions do not mean that we collect your fingerprints or biometric templates.
6. Security
We use HTTPS for app/backend communication, authenticated ownership checks for document access and private cloud storage. Backend credentials are not provided to the client. These safeguards reduce risk but cannot guarantee absolute security. Keep your sign-in account, password and downloaded documents secure.
7. Intended audience
The service is intended for adults aged 18 and over, including office workers, teachers and university students. It is not designed or marketed for children. This intended-audience statement is not a claim that this build verifies ages or blocks all under-18 access.
8. Updates and contact
We may update this policy as the service changes and will update the date above. For privacy questions, access/correction requests, deletion requests or concerns about a document, contact Sunlight Rose Studio at nhat.ngtan@gmail.com.
Provider information: https://firebase.google.com/support/privacy • https://cloud.google.com/privacy • https://www.netlify.com/privacy/ • https://developers.openai.com/api/docs/guides/your-data