Privacy Policy

Effective date: October 3, 2026 (amended October 25, 2026: completed the list of tools that download AI models)

FreeSoundTools (the "Site") explains here how personal information is handled, in line with Article 30 of the Personal Information Protection Act of the Republic of Korea and Article 31 of its Enforcement Decree. The Site has no sign-up or login, and does not send the files you add or your microphone audio to a server. One exception: in Text to Speech, if you pick a device voice marked "online", the text you type may be sent to the browser or operating system provider's servers (section 7).

1. Purpose of processing

The Site operator does not directly collect or store users' personal information. The third-party services described in sections 3, 7 and 11 may process connection data to deliver the website, keep it secure and show ads.

2. Information processed

CategoryItemsProcessed by
Collected directly by the Site operatorNone-
Generated automatically when you visitIP address, time of access, requested URL, browser and device information, referring pageHosting, CDN and model providers (section 7)
When you view a page with ads (guides, about, updates)Cookies, advertising identifiers, IP address, browser and device information, pages visitedAd providers (section 11)
Files or audio you add or recordNot processed by us (handled only in your browser, never sent to a server)-

The Site does not set cookies to identify you. Speech recognition models are kept in your browser's storage (Cache Storage). This storage is never sent to the operator's server, and you can clear it with your browser's "clear site data" option.

3. Retention period

The Site operator does not keep personal information. Connection data processed by third parties is kept according to each provider's policy (links in section 7).

4. Deletion

Because the operator holds no personal information, there is no separate deletion procedure. You can delete models and cookies stored in your browser yourself in your browser settings. You can also see and delete downloaded AI models one by one on the model manager.

5. Sharing with third parties

The Site does not provide personal information to third parties.

6. Service providers

ProviderService
Cloudflare, Inc.Website hosting, content delivery, security

7. International transfers

When you use the Site, your browser connects directly to the servers of the overseas providers below, and connection data such as your IP address is passed to them. Your file contents are not.

RecipientCountryItemsPurposeWhen and howRetention
Cloudflare, Inc. (https://www.cloudflare.com/privacypolicy/)United States (stored mainly in the US and EEA, accessible from locations worldwide)IP address, time of access, requested URL, browser and device informationWebsite delivery and securitySent automatically over the network when you visitPer the provider's policy
Hugging Face, Inc. (https://huggingface.co/privacy)United States (servers), with an entity in FranceIP address, time of access, browser and device informationProviding AI model files (fallback). Model files are first downloaded from the Dagochim family model storage (models.dagotools.com, served via the Cloudflare hosting above); only if that fails are they fetched from the original distributor, Hugging FaceOnly when an AI tool first downloads its model and our storage can't be reached (later the copy stored in your browser is used)Per the provider's policy
Google LLC and other ad providersUnited States and othersCookies, advertising identifiers, IP addressShowing and measuring ads, preventing invalid clicksWhen a page with ads loads, your browser connects directlyPer the provider's policy

How to opt out: if you don't use the tools that need these external connections (Audio to Text, Subtitle Generator, AI Meeting Notes and the High-quality AI voice in Text to Speech on this English site, plus English pronunciation practice and AI English conversation practice on our Korean site), no model or runtime requests are sent to those providers. Connecting to the Site itself requires hosting and can't be refused. In Text to Speech, choosing a voice marked "online" may send your text to the browser provider's servers. That processing follows the policy of the browser or operating system provider.

8. Children

The Site has no sign-up and does not collect personal information, so it has no procedure for obtaining a legal guardian's consent to process the personal information of children under 14. We do not show personalized ads based on the activity of users we know to be under 13. The Site is not directed primarily at children.

9. Your rights

You can ask to access, correct, delete or stop the processing of your personal information. Because the operator holds no personal information, please send requests about information held by third parties to those providers. To contact the operator, use the address below.

10. Security measures

11. Cookies, analytics and ads

Tool error statistics

Since 2 November 2026, when a tool hits an error we collect only the following, anonymously, so we can fix it: the page path (without query string or #), the error type with the FreeSoundTools script file name and line number where it happened, and the browser family (for example "Chrome, mobile"). File names, file contents, text you typed, the error message itself and your IP address are never sent or stored. The receiving server (ttdu-errlog.milkydotstudio.workers.dev, Cloudflare) keeps only daily counts, and a page sends at most 5 reports. Errors from scripts of other sites, such as ads or browser extensions, are not sent.

Anonymous usage counts

Since 8 November 2026, to learn which tools are actually used, when a tool creates a result file for you to download we count only: the site, the page path (no query or # part), the tool identifier (ID) and a marker whose value is "done" (completed). The receiving server adds the date (per day). File names, contents and sizes, anything you typed, your IP address, cookies and any value that identifies a person or device are never sent or stored. Each tool is counted at most once per browser tab; to do that we keep a marker only in this tab, which is removed when you close it. The receiving server (ttdu-errlog.milkydotstudio.workers.dev, Cloudflare) keeps only counts per date, site, page and tool.

Change history: 8 November 2026 — added "Anonymous usage counts".

12. Privacy officer and contact

13. Remedies

If you need help with a privacy violation, you can contact the Personal Information Dispute Mediation Committee (https://www.kopico.go.kr), the KISA Personal Information Infringement Report Center (https://privacy.kisa.or.kr), the Supreme Prosecutors' Office or the Korean National Police Agency's Cyber Bureau.

14. Changes to this policy

If this policy changes, we will post the effective date and the changes on this page.

Effective dateChange
2026-10-25Completed the list of tools that download AI models in the opt-out paragraph of section 7 (Audio to Text, Subtitle Generator, Meeting Notes, High-quality AI voice in Text to Speech, English pronunciation practice, AI English conversation practice)
2026-10-13Google AdSense ads now run only on the guide, English home, about and updates pages (not on tool pages or the Korean home with file drop); ad items in sections 2, 7 and 11 and the EEA consent message updated
2026-10-06AI model files now come first from the Dagochim family model storage (models.dagotools.com), with Hugging Face only as a fallback
2026-10-06Fonts and AI runtime files (ONNX Runtime, Transformers.js) are now served from this site, so jsDelivr was removed. Only AI model files come from Hugging Face
2026-10-04Added the use of a visit statistics tool (Cloudflare Web Analytics)
2026-10-03First version

Sources

Operator information

Operator: 행복유통
Representative: 박지훈
Business registration no. 280-30-01413
Address: 경상남도 사천시 사남면 병둔2길 70 (Republic of Korea)
Phone: +82-10-2101-6758
Email: [email protected]
All tools on this site are free and nothing is sold, so no mail-order business registration number applies.