VlogMe
EN
EnglishРусский日本語简体中文
Get started
VlogMe

Explore VlogMe

From your first idea to a finished video.

Get started

Make a photo talk with AI

To make a photo talk, upload one clear portrait, add the words as typed text or your own recording, and VlogMe animates the face so it speaks them in a lip-synced video you preview and download.

  • Your own photo, not a stock presenter
  • Typed words or your own recording
  • Credit quote before each render
Create talking photo

Try it with your own files

Set up your generation

Sign in to generate. This page opens Video Studio's Talking avatar workflow with Hedra Avatar selected. The optional field above is for delivery direction only; the spoken words are added in Video Studio.

  1. Start with a clear portrait
  2. Add a recording or typed script
  3. Set the model and framing
  4. Check the quote, then generate

01Talking photo examples

Five photos people bring, and what each one says

Each card names the photo to bring and the line that suits it. Use only photos and voices you have the right to use.

  • A still selfie turned into a talking video: a woman in denim overalls greets her followers
    Talking selfie9:16
    Source selfie: a woman in denim overalls under a thatched roof
    Photo

    Selfie

    A selfie that says hello

    Bring a front-facing selfie in good light with the mouth visible. Give it a greeting, an announcement or a quick tip, and post the vertical 9:16 result as a Story, Reel or Short.

  • Old family photo

    An old family photo that tells its story

    Photograph or scan the print straight on and crop it to one adult's face. Keep the line short and personal: a name, a year, a memory the family shares. Ask the family before you share it.

  • Talking pet: a grey tabby cat on a windowsill denies knocking over the plant
    Talking pet9:16
    Source still: a grey tabby cat on a sunny windowsill above a tipped-over plant
    Photo

    Pet

    A pet with something to say

    Choose a photo where the eyes and mouth are clear and the head faces the camera. Give the pet a voice and a one-line excuse; more ideas are on the talking animal page.

  • Team headshot

    A team headshot that welcomes new hires

    Use the company headshot of a colleague who has agreed to it. A short welcome, an onboarding step or a weekly update in landscape 16:9 fits an intranet page or a slide.

  • Memorial photo

    A memorial photo for a tribute

    Choose a portrait the family agrees on, and use a recording of the person's own voice that the family holds, or words the family writes. Show it privately first, and say that it was made with AI when you share it.

02Voice, price, limits and consent

What to know before the first render

Lip-synced talking portrait: a man in glasses and a vest delivers a quick line
Lip-sync9:16

01Typed speech or uploaded audio

Type the words, or upload your own voice

The video follows a speech track, and you choose where it comes from.

Upload audio when you already have the voice: your own recording, a voice note or a narration as an MP3, M4A, WAV or OGG file, or a voice track saved in Lab. Nothing is generated for the voice, so you go straight to the video quote, and the delivery is exactly yours.

Type the words when you do not: open “No voice track yet? Type the words”, enter up to 5,000 characters in “Words to speak”, pick a voice and press “Make the voice track”. The voice track is quoted first; once it is ready, it becomes the speech track for the render.

Either way, the words come from the audio. The optional “How should they say it?” field only shapes delivery: tone, gaze, expression.

Try it with your own photo
Vertical social clip: a woman with coffee and flowers shares a morning ritual
Ready for social9:16

02Price

What a talking photo costs

A talking photo is priced in credits by the length of its speech track, and Video Studio shows the exact quote before you confirm each render. A typed script adds a separately quoted voice track; your own recording adds no voice charge.

Planning in Director is free to start, with its images, voice and music, within usage limits, but a talking-photo render needs credits: buy a one-time credit pack, whose credits never expire, or choose an optional plan. See pricing for current packs and plans.

Try it with your own photo
A presenter in a green blouse introduces a garden workshop
More than a talking head16:9

03Limits

What works, and what does not

A video is as long as its speech track: up to 10 minutes with Hedra Avatar and up to 60 seconds with OmniHuman 1.5.

One clear, front-facing face with the mouth visible works best. Sunglasses, a hand over the mouth, a profile view or a very small face in a group photo sync less cleanly, so crop a group photo to one person first. Drawings, paintings and mascots work when the eyes and mouth are clearly drawn.

Hedra Avatar renders 720p in 9:16, 16:9 or 1:1 and moves the face and head; OmniHuman 1.5 keeps your photo's framing and adds gestures. To animate a whole scene instead, use image to video.

Try it with your own photo

04Consent

Only photos and voices you may use

You need the rights to the portrait and the voice before any render; with Hedra Avatar you confirm them in Video Studio. For anyone other than yourself, that means their permission; for a person who has died, the family's. Do not impersonate anyone, do not animate public figures or other people's photos without a release, and never use the face or voice of anyone under 18.

VlogMe adds no visible watermark, so the label is yours to add: when you publish, say that the video was made with AI wherever the platform asks.

Try it with your own photo

Video StudioAI avatar generator

Every talking-avatar tool in one place
  1. Your starting materialA portrait and the words to speak
  2. The resultA lip-synced talking photo video

Use an existing recording or generate a voice from text. Voice generation and the avatar render have separate quotes; editing, dubbing and publishing open separate workflows.

Continue in Video Studio

03Step by step

How to make a photo talk in Video Studio

  1. Start with a clear portrait

    Choose a photo where the face is well lit and facing the camera, with the mouth visible. JPG, PNG and WebP are accepted. In Video Studio, upload the portrait or pick one already saved in Lab.

  2. Add the voice, model and framing

    Upload a recording or choose one from Lab, or type the words and generate a separately quoted voice track. Keep Hedra Avatar to animate your own portrait, choose 9:16, 16:9 or 1:1, and confirm your rights to the portrait and voice.

  3. Check the quote, generate, preview, download

    Press “Create avatar video”. Video Studio shows the exact credit quote before you confirm this video render. Confirm, wait for the job, then preview the result in the Studio player and use “Download” to save it. “Generate again” requests another version and is quoted again.

  • Script → AI voice
  • Uploaded audio
  • 9:16
  • 16:9
  • 1:1

Finished examples

A craftsman, a sushi mascot, an ad

A woodworker in a leather apron talks about his workshop
9:16
Source portrait: a woodworker in a leather apron
Source image
An animated sushi character promotes itself in a playful voice
9:16
Source image: a 3D salmon sushi character
Source image
Social ad: a woman presents an unlabeled face mist in three seconds
9:16
Source portrait: a woman holding an unlabeled spray bottle
Source image

Your finished clip

Preview, download, and Lab

When the job completes, the clip plays in the Studio player. From there you can download the video — Hedra Avatar renders a 720p MP4 with no visible watermark — request another version, or open the result in Lab. Generated voice tracks and videos are saved to Lab, where you can find them for the next version.

The result card also offers entry points to Edit video (trim, crop and on-screen text in Video Editor), Dub this video, Publish, and Edit with Aisma. Each opens its own workflow with its own steps.

  • Preview in the Studio player
  • Download the finished video
  • Find voice tracks and videos in Lab
  • Separate editing, dubbing and publishing workflows
Create talking photo

Talking photo FAQ

Questions before you create

Is it free?

You can start free in Director: the script, images, voice-over and music cost nothing there, within usage limits. A talking-photo render needs credits, from a one-time credit pack or an optional plan, and Video Studio shows the exact credit quote before you confirm a render.

How do I make a photo talk?

Upload a clear portrait in Video Studio, add a recording or type the words for an AI voice, choose the framing, check the credit quote and press “Create avatar video”. The finished clip plays in the Studio player, ready to download.

Can I use my own voice?

Yes. Upload a recording as an MP3, M4A, WAV or OGG file, or pick a voice track from Lab. Your recording is used as it is, and no voice generation is charged.

What photo works best?

A clear, front-facing portrait with good light and the mouth visible, in JPG, PNG or WebP. Crop a group photo to one face first.

How long can a talking photo be?

As long as its speech track: up to 10 minutes per video with Hedra Avatar and up to 60 seconds with OmniHuman 1.5. With Hedra Avatar you choose 9:16, 16:9 or 1:1 framing.

Is there a watermark?

No. VlogMe adds no visible watermark to the video you download. When you publish it, say that it was made with AI wherever the platform asks.

Can I make an old or black-and-white photo talk?

Yes, if the face is an adult's, clear and seen from the front; never use the face of anyone under 18, even in an old photo. Photograph or scan the print straight on, crop it to one face and keep the first line short. A photo of someone who has died needs the family's agreement.

Can I make a drawing, painting or pet talk?

Yes, when the face has clearly drawn eyes and a mouth. Drawings, paintings, mascots and pets go through the same steps as a photo; the sushi mascot on this page started as one generated image. For pets, see talking animal.

Can I make someone else's photo talk?

Only with their permission, or the family's for a person who has died. With Hedra Avatar you confirm the rights to both the portrait and the voice before rendering. Do not impersonate anyone, do not animate public figures without a release, and never use the face or voice of anyone under 18.

Which model should I choose?

Keep Hedra Avatar for a portrait you upload: it renders the longest videos and lets you pick the framing. OmniHuman 1.5 also uses your photo and adds gestures for clips up to 60 seconds. HeyGen Avatar V renders a fixed presenter, not your image.

Can the photo sing instead of talk?

Yes, with a song instead of speech: see make a photo sing, where OmniHuman 1.5 follows the vocals with gestures.

Can I change the words or edit the video afterwards?

Changing the words means a new speech track — upload a different recording or generate a new voice track — and another quoted render. For the existing video, Edit video opens Video Editor for trimming, cropping and on-screen text; captions are created in Audio Studio, and publishing runs through the Social handoff. Each is a separate step.

Tool

Make a photo sing

Give the same portrait a song instead of a speech, with gestures that follow the vocals.

Tool

AI avatar generator

Explore the AI avatar generator and available models.

Related tools

Models

Key terms

Compare

From the blog

Ready when you are

Make the first version, then make it yours

Pick a portrait, add the words, and see the quote before each paid generation.

Create talking photoSee pricing

Your portrait · Your recording or an AI voice · A quote before each paid generation