How to lip sync a video with Magic Hour: 5 steps

Quick answer

To lip sync a video in Magic Hour, upload footage with one visible face, add the approved speech or song audio, generate, then review the mouth during speech, pauses and head turns. The current guest tool provides three attempts per day, each up to ten seconds, with a watermark. If you only have a portrait, use Talking Photo instead. Limits checked September 13, 2026.

Lip-sync a video with AI

Upload a video and audio track, then review timing and mouth movement in Magic Hour.

Try AI Lip Sync

Magic Hour publishes this guide and includes its own product in the comparison. Treat the recommendations as editorial guidance from a vendor, and verify the linked first-party product details and your own output requirements before choosing a tool.

Quick Comparison

You have	Use	Guest access
Video plus an audio track	Magic Hour Lip Sync	Three attempts/day, up to ten seconds, with a watermark
Portrait plus an audio track	Talking Photo	Three attempts/day; the FAQ states a five-second maximum; free video may be watermarked
A script without recorded speech	Create and review the audio first	Voice-generation access is separate from the lip sync allowance

What Is AI Lip Sync?

AI lip sync makes a person in a video appear to speak or sing along to completely new audio.

The AI analyzes facial movements, especially around the mouth, and adjusts them so they match the uploaded sound.

Lip sync can replace dialogue or synchronize a visible performance to another audio track. For translation, approve the translated meaning and pronunciation before generating the synchronized video.

Lip Sync vs Talking Photo: What Is the Difference?

Lip sync works with an existing video where a person is already moving and speaking on screen. A talking photo tool starts with a still image and animates it so the subject appears to talk or sing. If you already have a video, use lip sync. If you only have a photo, use Talking Photo instead.

Use the input you actually have. Rebuilding an existing performance from a still image can discard the original head movement and camera motion.

If you have a still photo instead of a video, Magic Hour's Talking Photo tool is what you need.

What You Need Before You Start

Before creating your first lip synced video, gather the following:

A short MP4 or MOV video with one clearly visible face
A clear audio recording of the line or song, with similar duration
Permission to use the video, voice, and likeness for your project
A browser; no account is required for the guest trial

The free video trial is limited to ten seconds. Start with a representative sample before choosing a longer account-based job.

How to Lip Sync a Video With AI: Step by Step

Step 1: Go to Magic Hour's Lip Sync Tool

Open Magic Hour's Lip Sync tool in your browser. The tool loads directly from the web, so there is nothing to install.

lipsync

Step 2: Upload Your Video

Click the upload button and choose the video you want to modify. Videos where the subject is clearly facing the camera generally produce the most natural-looking results.

Step 3: Upload Your Audio

Upload the approved audio track. For a first trial, record “Pick the blue box. Pause. Now press play.” Include the pause in the recording and keep the line within ten seconds. This gives you a concrete moment to inspect for unwanted mouth movement during silence.

Step 4: Click Generate

Select the generation button shown in the tool and allow the job to finish. Keep the page or saved project available so you can return to the result. Processing time varies with the input, settings, and queue.

Step 5: Review and Download

Watch the complete output with sound, then inspect pauses, head turns, teeth, and face edges. Download the sample and confirm its duration, crop, and watermark before generating a longer version.

Tips for the Best Results

Keep the face unobstructed and in frame.
Use a clear recording with one speaker for the first attempt.
Check audio and video start times if the entire track is consistently early or late.
Keep intentional pauses; remove unwanted leading silence only when it causes a timing mismatch.
Change the input or setting related to a visible failure before spending another attempt.

Common Mistakes New Lip Sync Users Make

Using low-resolution videos. Blurry footage gives the AI less facial detail to work with, which can reduce lip sync accuracy.
Uploading videos where the person's face is partially hidden. Sunglasses, hands, microphones, or objects covering the mouth can make synchronization less reliable.
Choosing clips with heavy motion blur or fast camera movement. Stable footage usually produces better results.
Using audio with loud background noise. Clear speech or vocals help the AI match mouth movements more accurately.
Expecting AI to fix poor source material. Even the best lip sync tools perform better when both the video and audio are clean.

What Can You Use AI Lip Sync For?

Dubbing videos into different languages while matching mouth movements.
Making a person appear to sing along to a song.
Creating funny social media clips and meme videos.
Syncing voiceovers to silent footage.
Producing content without needing to reshoot video.

For an explainer or product video, check the surrounding edit too. A corrected voice line does not update an outdated caption, label, or screen recording.

What to Try Next

If the sample meets your requirements, check the selected mode, duration, and credit quote before making the full version. Paid plans provide watermark-free video and commercial use under the terms. For cost examples and alternatives, see the lip sync tool comparison.

Continue learning

Continue learning: AI lip sync explainer, realistic talking-avatar guide, and best talking-photo tools.

Frequently Asked Questions

Yes. As checked September 13, 2026, the guest tool offers three attempts per day without signup, up to ten seconds each, with watermarked free video. Account credits and rewards are separate; check the current quote before a longer job.

Magic Hour Lip Sync supports MP4 and MOV files, which cover most videos exported from phones, cameras, and editing software. If your file is in another format, converting it to MP4 is usually the easiest option before uploading.

Use a common audio file such as MP3 or WAV and check the upload control for its accepted formats. The documentation lists additional supported audio formats.

The current guest video limit is ten seconds. For longer work, check the limits and quote for the selected mode in your account before uploading the whole project.

Results depend on the source and settings. Review the full clip rather than assuming successful generation means the mouth, identity, and motion are correct.

Yes, you can supply audio in a different language. Lip alignment does not verify translation accuracy. Have a fluent speaker review the script and voice before generating the final video.

For the mechanism, evaluation signals and failure modes behind the workflow, read what AI lip sync is.