No subscription

One photo, speaking your words

Upload a portrait, write the words, pick a voice, and HeyGen Avatar IV animates the face to say them, lip-synced, for as long as the script runs: you pay for the seconds it takes to say. For a short line inside a scene, with the room's sound, Veo 3.1 Lite and Wan 3.0 animate the photo from a prompt. Use photos of yourself or of people who agreed to it.

Try it with an example See what a clip costs

$0.50 of free credit when you verify your email. Clips from $0.30, top-ups from $5, and top-up credit never expires.

Diese Seite auf Deutsch

An example

One photo and one script, and the clip it made

An AI-made portrait of a woman holding a small amber bottle of facial serum — The photo that went in

“Hi, I'm here to show you the Luma facial serum. Three drops in the morning, three at night.”

The script, said word for word

The clip that came back

Made with HeyGen Avatar IV on November 10, 2026, from this one AI-made portrait, this script and the voice Cassidy: a 4.6-second clip, ready in about a minute and charged for its 4.6 seconds. The person does not exist, and Luma is a made-up brand.

What a clip costs

Which model to use

Today’s price for a typical clip on each model, from GenMagic’s live price list. You pay only for the clips you make.

Model	Why for this	A clip	Longest take	Up to
Avatar IVHeyGen without a subscription	your script word for word in a voice you pick, any length, billed per second of speech	$0.385 s of speech in 720p	The script's length	1080p
Veo 3.1 LiteVeo 3.1 without a subscription	a short spoken line inside a scene, with sound, for a low price	$0.304 s in 720p	8 s	1080p
Wan 3.0Wan 3.0 without a subscription	a line with gestures and room sound, up to 30 seconds in one take	$0.325 s in 480p	30 s	1080p

Start from an example

Prompts that work

Each opens the studio with the prompt and the model ready. Change it, see the price, generate.

The example above, ready to run

Hi, I'm here to show you the Luma facial serum. Three drops in the morning, three at night.

Avatar IV, $0.38 a clipOpen in the studio

A line inside a scene

Animate this photo: the woman looks into the camera, smiles and says, "Hi, I'm Lena from the bakery on Main Street. Come by before ten and the coffee is on us." Warm, friendly voice, subtle head movement, soft bakery sounds, 8 seconds. Keep her face exactly as in the photo.

Veo 3.1 Lite, $0.30 a clipOpen in the studio

A longer message with gestures

Animate this photo: the man in the workshop looks into the camera and says, "Welcome to our workshop. Everything you see here is made by hand, from the first cut to the last coat of oil. Ask us anything." Calm, deep voice, natural gestures, quiet room tone, 12 seconds. Keep his face exactly as in the photo.

Wan 3.0, $0.32 a clipOpen in the studio

What makes the difference

How to get good results

Use a sharp, front-facing portrait with the mouth visible and even light; sunglasses and hands over the face confuse the lip sync.
On Avatar IV the prompt is the script: write only the words to be said, and pick the voice in the studio's settings. Kore and Orus speak any language; Cassidy, Andrew, Leonie, Otto and the others speak one.
On Veo 3.1 and Wan 3.0, put the exact words in quotes and describe the voice (warm, deep, a slight accent): the model makes the voice from that, and each take runs up to 8 or 30 seconds.
Only animate photos of yourself or of people who agreed to it, and never present an AI person as a real customer giving a testimonial.

Go deeper: AI avatar videos from one photo, Spoken video ads from a product photo, Any picture to video, HeyGen without a subscription, A D-ID alternative without a plan.

Questions

How do I make a photo talk?

Open the studio on HeyGen Avatar IV, attach the portrait, write the words it should say and pick a voice. The clip is the person in the photo saying your script, lip-synced, as long as the words take.

Can it speak other languages?

Yes. Avatar IV has voices for English, German, Spanish, French and Japanese, and two multilingual voices (Kore and Orus) that read a script in any language. On Veo 3.1 and Wan 3.0, write the line in the language you want and test it on a short clip first.

What does a talking photo cost?

Avatar IV is billed per second of speech, so a script costs what it takes to say; the studio shows the price for your script before you generate. The table on this page shows today's prices on each model.

Can the same voice speak in every clip?

Yes, on Avatar IV: pick the voice by name, and every clip you make with it speaks in that voice.

Do I need a subscription?

No. You top up a balance and pay per clip; top-up credit never expires.

What does a clip cost?

Today a typical clip costs $0.38 on Avatar IV (5 s of speech in 720p), $0.30 on Veo 3.1 Lite (4 s in 720p), $0.32 on Wan 3.0 (5 s in 480p). The studio shows the exact price for your settings before you generate, and a clip that fails is not charged.

Make your first clip on Avatar IV

$0.50 of free credit to start. No subscription, no card to try.

Open the studio