LongCat Avatar 1.6 AI Lip Sync With Up to 64s Audio

LongCat Avatar 1.6 generates lip-synced video from a reference image and up to 64 seconds of audio. Supports single speakers and two-person dialogue at 480p or 720p.

LongCat Avatar 1.6 Studio

Generate a single-speaker or two-speaker lip sync video from a reference image and recorded audio.

Preview
Demo, generation progress, and final output
Two-person lip sync preview

Play the sample to see how lip movements and expressions follow each speaker’s lines.

What Is LongCat Avatar 1.6?

LongCat Avatar 1.6 turns a reference image and recorded audio into a lip-synced avatar video, with up to 64 seconds of audio per generation.

Reference image
A portrait supplies the speaker’s face and framing. For two speakers, use an image with both people.
Voice recording
Add one recording for a single speaker, or a separate track for each person in Multi Avatar.
Optional scene prompt
Describe the expression, framing, and gestures to guide the video.
Singing performance
Singing performance

LongCat Avatar 1.6 Features for Audio-Driven Lip Sync

Animate one speaker or two, guide the scene with an optional prompt, and choose 480p or 720p output.

  • Indoor speaking shot
    Indoor speaking shot

    More controlled mouth movement

    Recorded speech drives the avatar’s mouth movements and expressions. The reference image supplies the face and framing, while an optional prompt describes the scene and movement. Preview the result with sound to review speech timing.

  • Outdoor speaking shot
    Outdoor speaking shot

    Up to 64 seconds of audio

    Each generation supports up to 64 seconds of audio at 480p or 720p. Choose a portrait that shows the speaker’s face clearly, then describe the intended expression or gesture in the optional prompt.

  • Outdoor dialogue
    Outdoor dialogue

    Separate tracks for a dialogue scene

    Multi Avatar uses an image containing both speakers and a separate recording for each person. Choose Simultaneously, Left then Right, or Right then Left to control the audio order. Simultaneous playback uses the longer track for billing, while sequential playback uses their combined duration.

What Can You Create With LongCat Avatar 1.6?

Create presenter clips, dialogue between characters, and short vocal performances.

  • Training and lesson videos

    Pair a presenter portrait with a lesson excerpt of up to 64 seconds. A crop showing the face and upper body makes the speaker’s mouth movements and gestures visible. Download the finished clip to use alongside slides or a screen recording.

  • Dialogue scenes

    Set up a supported multi-person reference and separate the two voice tracks. Start with alternating lines and look for clean speaker changes.

  • Singing and character animation

    Try a short performance using a recording you can use. Watch how the expression changes over a held note or a pause.

Watch LongCat Avatar 1.6 Video Examples

Compare lip timing, facial expressions, and speaker changes in the example videos.

Who Is This Audio-Driven Avatar Workflow For?

Use a portrait and recorded voice for course introductions, product explanations, and character dialogue.

  • An instructor recording a lesson beside printed course notes

    Course creators

    Test an opening that includes pauses and technical terms. Keep the presenter in a medium crop so you can inspect the mouth and still see the upper body.

  • A trainer demonstrating a handheld device to a colleague

    Product and training teams

    Create the speaking portion around a recording approved for the project. Edit slides or screen recordings separately where the viewer needs to see a process or interface.

  • An animator working with sketches and a small clay character

    Animation teams

    Try a short line of dialogue or a vocal hook using your own art. Compare subtle speech with more expressive delivery before you choose a shot style for the rest of the scene.

How to Use LongCat Avatar 1.6 With a Photo and Audio

Write a prompt, add a reference image and audio, then save an MP4 with LongCat Avatar 1.6.

  1. Step 1. Write a prompt

    Use the Prompt field to describe the scene, framing, and how the avatar should move. For a talking avatar, try “A presenter speaks by a window, pausing between sentences.” The audio supplies the words and their timing.

    An active Prompt field with a scene description and a text cursor
  2. Step 2. Upload your image and audio

    Upload a portrait and one voice recording for Single Avatar. For Multi Avatar, use an image with both people and a separate recording for each speaker. Choose the speaking order and 480p or 720p output, then select Generate Avatar Video. The generation limit is 64 seconds.

    The Upload media area with a portrait image and a voice recording
  3. Step 3. Download and save your avatar video

    Play the finished video in Preview and check that the mouth movements follow the voice. Select Download to save the MP4 for editing, sharing, or adding to a lesson.

    A finished talking avatar preview with a Download video button and an MP4 filename

LongCat Avatar 1.6 vs 1.5 vs 1.0

Compare the audio limits, generation costs, inputs, and speaker modes available on this site.

ModelAudio durationResolution and creditsRequired inputsSpeaker modes
LongCat Avatar 1.664s480p at 2 credits/sec · 720p at 4 credits/secReference image and one or two audio tracks. Optional prompt.Single speaker or two speakers. Simultaneous and sequential playback.
LongCat Avatar 1.530s480p at 1 credit/sec · 720p at 2 credits/secReference image and one or two audio tracks. Optional prompt.Single speaker or two speakers. Sequential playback.
LongCat Avatar 1.060s480p at 1 credit/sec · 720p at 2 credits/sec · 1080p at 3 credits/secReference image and one MP3 recording. Optional prompt.Single-speaker talking avatar.
The same portrait in LongCat Avatar 1.5 and 1.6
The same portrait in LongCat Avatar 1.5 and 1.6
LongCat Avatar 1.6 Pricing

Choose Your Credit Pack

One-time credit packs for LongCat Avatar 1.6. Generation costs depend on the audio duration and selected resolution. Credits never expire.

Base

$9.9one-time
Get 90 Credits
90 Credits
Up to 64s audio per generation
3s minimum billed duration
480p at 2 credits/sec
720p at 4 credits/sec
Single speaker and two-speaker modes
Simultaneous or sequential dialogue
MP4 video downloads
Most Popular

Pro

$29.9one-time
Get 400 Credits
400 Credits
Up to 64s audio per generation
3s minimum billed duration
480p at 2 credits/sec
720p at 4 credits/sec
Single speaker and two-speaker modes
Simultaneous or sequential dialogue
MP4 video downloads

Ultimate

$49.9one-time
Get 800 Credits
800 Credits
Up to 64s audio per generation
3s minimum billed duration
480p at 2 credits/sec
720p at 4 credits/sec
Single speaker and two-speaker modes
Simultaneous or sequential dialogue
MP4 video downloads

Creator

$99.9one-time
Get 1800 Credits
1800 Credits
Up to 64s audio per generation
3s minimum billed duration
480p at 2 credits/sec
720p at 4 credits/sec
Single speaker and two-speaker modes
Simultaneous or sequential dialogue
MP4 video downloads

Choose one-time credits • Flexible billing options

✓Choose one-time✓Credits never expire✓Secure payments

LongCat Avatar 1.6 FAQ

Find the image and audio requirements, duration limit, playback options, and generation costs.

Can I create a talking avatar from audio and a still photo?

Yes. Upload a reference image and a voice recording, then select Generate Avatar Video. The image can be JPG, PNG, WebP, GIF, or AVIF, up to 10 MB. Audio can be MP3, WAV, M4A, OGG, or FLAC, up to 20 MB. A clear face and a recording with distinct words and pauses help you review the resulting lip sync.

What does the optional prompt control?

The prompt describes the scene, framing, expression, or movement you want. For example, “A presenter speaks calmly by a window with gentle hand gestures.” A reference image and audio are required in both Single Avatar and Multi Avatar modes. You can leave the prompt empty.

How long can a LongCat Avatar 1.6 generation be?

Each generation supports up to 64 seconds of audio. Audio exceeding this limit is clipped for generation. Billing uses whole seconds with a minimum of 3 seconds and a maximum of 64 seconds per task. Two-speaker playback order determines how the two audio durations are combined.

Can two people speak in the same scene?

Yes. Select Multi Avatar, upload an image with both people, and assign each recording to Left Speaker Audio or Right Speaker Audio. Simultaneously plays the tracks together and bills for the longer track. Left then Right and Right then Left play them in sequence and bill for their combined duration, within the task’s duration limits.

How much does LongCat Avatar 1.6 cost?

Generation costs 2 credits per second at 480p or 4 credits per second at 720p. The 3-second minimum costs 6 credits at 480p or 12 credits at 720p. A 64-second task costs 128 or 256 credits respectively. The Generate Avatar Video button shows the estimate for your inputs. Credit packs are one-time purchases.

What changes when I switch between 1.6 and 1.5?

Version 1.6 supports up to 64 seconds of audio and includes simultaneous two-speaker playback. Its rates are 2 credits per second at 480p and 4 at 720p. Version 1.5 supports up to 30 seconds and sequential two-speaker playback, at 1 credit per second at 480p and 2 at 720p. The Model selector updates the available controls and credit estimate.

Open the LongCat Avatar Generator

LongCat Avatar 1.6

Prepare your first talking avatar video

Choose a portrait and add a voice recording for a presenter or a dialogue scene.