1. What data we process
- Original sound file:Audio/video clips you record via microphone or upload from your device.
- Sound characteristics:Timbre, pitch, rhythm, pronunciation, voiceprint characteristics, quality inspection and usability results.
- Sound identification:Third-party generated voice ID, model task ID, status, and error information.
- Verify and generate content:Test text, audio and final results used to verify sounds or generate songs.
- Operation record:Account, authorized version, time, source, device and necessary security logs.
2. Processing process
Only after you check this authorization and actively click record or upload, the browser will request microphone permission or read the file you selected. Browser permissions are independent of this authorization. You can revoke microphone permissions in the system or browser.
3. Purpose and necessity of processing
We only process sound data for the following purposes: checking audio quality and format, confirming that sound functions are available, creating or calling sounds of your choice, generating songs, troubleshooting failed tasks, and ensuring account and service security.
Sound functionality cannot be provided without processing sound samples and characteristics at all. You can deny or withdraw authorization and continue to use normal AI music generation functionality that does not require sound samples.
4. Confirmation you make when checking
By checking this authorization, you confirm:
- You are over 18 years old and upload your own voice; or you have obtained verifiable, clear and applicable authorization from the voice subject for AI processing and song generation;
- The audio does not contain unsolicited third-party voices, nor does it contain the voices of public figures, entertainers or minors;
- You will not use generated voices to impersonate others, defraud, harass, create false endorsements, or infringe personality, performance, or intellectual property rights;
- You have read the Privacy Policy, Terms of Service and this authorization, and understand the risks and withdrawal methods listed below.
5. Possible risks
- Voice characteristics have strong identity correlation and are not as easy to change as passwords after being leaked;
- The AI results may be different from the original sound, or may be mistaken by the audience for real singing;
- The service provider may fail to recognize, produce artifacts, or generate inappropriate content;
- After being shared or downloaded publicly, third parties may copy, forward or use it out of the original context;
- Different regions have different requirements for sound rights, performer rights, and AI logos.
6. Withdrawal of consent, deletion and appeal
You can stop using the sound function at any time, revoke browser microphone permissions, or contact us to withdraw this authorization and apply to delete sound samples and voice ID. Withdrawal will not affect the legality of previous processing based on consent, nor will it necessarily delete generated songs that you have actively disclosed or that a third party has legally saved.
We will verify the identity of the requester and process it within the scope of legal and technical permission; if we must retain some security, transaction or dispute records, we will explain the reasons.
Sound data request
Email subject suggestions:Voice Data Request / Voice Data Request