AI Lip Sync Documentation

Getting Started

1. Open the AI lip sync generator and choose Video or Photo. Local photo checks and sample selection are available in the released interface; the page shows whether configured remote photo generation is available.
2. Upload audio or record it in the browser. Enter Text is available only when text-to-speech is configured for the current deployment.
3. Confirm that you have permission to use the source, every person shown or heard, and the voice or text input.
4. Run the local preflight check before submitting a generation job.
5. When you are ready to generate, sign in, review the restored draft and credit estimate, then explicitly submit the job. Follow its progress from Activity.

Local preflight runs in your browser. The selected files are not uploaded just to perform that check.

Input Requirements

• Video: MP4, MOV, WebM, or M4V, up to 120 seconds.
• Audio: MP3, WAV, M4A/AAC, or WebM. Video lip-sync audio supports up to 120 seconds; Talking Photo audio supports up to 60 seconds.
• Avatar image: JPEG, PNG, WebP, AVIF, HEIC/HEIF, GIF, BMP, or TIFF. AVIF, HEIC/HEIF, GIF, BMP, and TIFF are converted in your browser to provider-safe PNG; animated GIFs use the first frame.
• A video of at least 480 × 360 pixels is recommended. Large differences between the audio and video duration may produce a warning.

For a complete preparation and review sequence, use How to Lip Sync a Video. For side profiles, occlusion, multiple people, motion, timing mismatch, blur, teeth artifacts, or unstable footage, use the Lip Sync Quality Guide.

Remote Generation

After the file check passes, sign in and submit a private generation job. Upload is available only when private storage is configured, and generation is available only for configured provider and quality combinations. Photo generation and text-to-speech are evaluated independently, so one may be available while the other is not.

The Pricing page shows the current catalog. Checkout is available only when the payment service is configured. A browser return page does not grant credits by itself; confirmed payment processing is the fulfillment source.

Generated media is kept account-scoped. Completed results generated with trial or purchased credits can be downloaded without a Visemix watermark from the generator or Activity. Review every result before publishing it, and disclose synthetic or altered media when required by law or by the platform where you publish it.

Rights and Consent

Only submit media that you own or are authorized to use. You are responsible for permission from the people shown or heard in the media and for any copyright, publicity, privacy, or other applicable rights.

Troubleshooting

• If a file is rejected, check its format, size, duration, and readable metadata.
• If generation fails, record the visible error message and send the job ID and reproduction steps through Support.
• If audio and video timing is noticeably different, use tracks with closer durations or review the warning before continuing.
• If a completed result is not available, wait for the job to finish and retry from Activity.
• If the same visual artifact remains after a meaningful input change, stop repeating the unchanged task and follow the quality troubleshooting guide.

Support

For account or job-specific help, use Support and include the job status, file types, and the steps that led to the problem. General questions can be sent to support@lipsync.vip. Do not send passwords, payment credentials, or private media by email. See the Privacy Policy for retention and deletion details.