How do I make a photo talk?

Upload a portrait, type what it should say, pick a voice and press generate. You get an MP4 of the person in the photo speaking your script, with sound.

Do I need a separate text-to-speech tool?

No. The voice is generated from your script as part of the same job, already synced to the lips. You do not upload any audio.

How much does it cost?

2 credits per second of speech, estimated from the script: 12 credits for the shortest clip and 60 for a 30-second one. Credits come with any NENE plan or credit pack.

Is the AI talking photo generator free?

No. Each video costs credits, shown before you generate. If a video fails, the credits are returned automatically.

Which languages can it speak?

English for now. Other languages will be added once they have been tested for natural speech and lip sync.

What kind of photo works best?

One person, facing the camera, face sharp and evenly lit, framed from the chest up, with nothing covering the mouth.

Can I use a photo of someone else?

Only an adult who agreed to be animated. Photos of children and public figures are not allowed, and neither are scripts meant to impersonate someone.

What is the difference between a talking photo and motion transfer?

A talking photo speaks your script with face and head motion. Motion Transfer AI makes your photo copy the body movements of a reference video, without a script.

Is there a watermark?

NENE does not add a watermark to the MP4 you download.

What happens to my photo?

It is used only to make your video. The prepared face copy is deleted when the video is finished; the MP4 stays in your history.

Photo + script → talking video

AI Talking Photo Generator

Upload one portrait, type what it should say and pick a voice. The AI reads your script out loud and animates the face to match: lips, blinks, expression and head movement. No separate voice-over needed.

01Photo

Click or dropFace visible · up to 5 MB

One person, facing the camera
Face clear, even light
Head and shoulders in frame
Your own photo, or one you have permission for

02What should it say?

Up to 30s of speech0/450

03Voice

This is my photo, or the adult in it agreed to be animated. It is not a public figure or a child, and the script does not impersonate anyone.

Voice built in · credits back if it fails

Talks

Your talking video appears here

The person in your photo reads your script out loud, with matching lip movement, blinks and head motion. The MP4 has sound.

How to make a photo talk (3 steps)

01
Upload a portrait
Add a JPEG, PNG or WebP up to 5 MB: one person, facing the camera, face clear and well lit. Head and shoulders work best.
02
Type the script and pick a voice
Write up to 450 characters in English and choose one of 8 voices. The cost updates as you type: 2 credits per second of speech.
03
Generate and download
Press generate. The photo is prepared, then the talking video is made with the voice already in it. Download the MP4 when it is ready; you can leave the page in the meantime.

Make your photo talk

What is an AI talking photo?

A talking photo is a still portrait turned into a short video of that person speaking. The AI generates a voice from your text, then animates the face so the lips follow every word, with natural blinks, small head turns and expression changes in between.

On NENE the whole job is one step. You upload a photo and type the script; the voice, the lip sync and the video are made together, and you download one MP4 with sound. Scripts are in English for now, up to 450 characters, which is about 30 seconds of speech.

How long can a talking photo be?

Speech runs at roughly 15 characters, or two and a half words, per second. The counter under the script box shows the estimate as you type, and that estimate is what you pay for. The shortest clip is billed as 6 seconds (12 credits), the longest is 30 seconds (60 credits).

Video length	Script	Credits
10 seconds	about 150 characters (25 words)	20
20 seconds	about 300 characters (50 words)	40
30 seconds	about 450 characters (75 words)	60

Need a longer talk? Split the script at the end of a sentence into 30-second parts, generate each part from the same photo and voice, and join the MP4s in any editor. Cutting at a full stop keeps the joins natural.

Voice built in: no separate text-to-speech step

Most photo-to-video tools only move the face. To make the photo speak you first create an audio file in a separate text-to-speech tool, then upload it. Here the script goes straight in: the voice is generated with the video, so the timing of the lips and the words already match.

Pick from 8 voices. They read English scripts naturally; punctuation shapes the delivery, so a comma gives a short pause and a full stop a longer one. Write numbers and abbreviations the way they should be said, for example "twenty twenty-six" rather than "2026".

Which photo makes the best talking head

Front-facing and sharp: The face is what gets animated, so it needs to be clear. A sharp photo looking at the camera gives the cleanest lip sync.
Even light, nothing over the face: Soft, even light works best. Sunglasses, a hand near the mouth, heavy filters or hair across the face make the motion less stable.
One person: Use a photo with a single person in it. With two faces the AI cannot know which one should speak.
Head and shoulders: A portrait framed from the chest up gives the face enough pixels. A full-body photo works, but the face is small and loses detail.

What people make with a talking photo

Spokesperson and product intros: Turn a team photo into a short talking intro for a landing page, a product update or an ad, and change the script without filming again.
Lessons and presentations: Teachers and course makers use an AI talking head to introduce a lesson, explain a slide or read a short summary in a consistent voice.
Social posts: A talking photo makes a quick hook for TikTok, Reels or Shorts: a greeting, a reaction or a one-line tip from your own portrait.
Characters and mascots: A human-like illustrated character or brand mascot with a clear face can talk too. Results are best when the face looks like a real one.

Talking photo vs motion transfer vs viral video remake

Compare	AI Talking Photo	Motion Transfer AI	Viral Video Remake
You upload	One photo + a script	One photo + a reference video	One photo + a reference video
What moves	Face and head, speaking	Your photo copies the body moves	You, inside the original scene
Sound	Yes, the voice reads your script	Not guaranteed	Not guaranteed

Want your photo to dance instead of talk? Use Motion Transfer AI. Want yourself inside a trending clip? Use Viral Video Remake.

Fixing common talking photo problems

The lips do not follow the words: Use a sharper, front-facing photo with the mouth fully visible. Small faces, side angles and objects near the mouth make lip sync less accurate.
A word is said wrong: Spell it the way it sounds. Write numbers, dates, prices and abbreviations out in words, and split long sentences with commas.
The video failed: Usually no clear face was found, or the photo or script did not pass the safety check. Your credits come back automatically; try a clearer photo.
The script is too long: The box takes up to 450 characters. Split longer scripts into parts and join the clips afterwards.

Credits and limits

The price is set by the length of your script and shown before you generate. Credits are taken when the job starts and returned automatically if it fails or has not finished after 2 hours.

Price: 2 credits per second of speech
Per video: 12–60 credits
Script: English, up to 450 characters
Length: 6–30 seconds (billed)
Photo: JPEG, PNG or WebP, up to 5 MB
Output: MP4 with sound

Credits come with any NENE plan or credit pack. See pricing. Want a bigger file afterwards? Run the MP4 through the Video Upscaler.

Rights and consent

Only animate your own photo, or a photo of an adult who agreed to it. Do not upload photos of children or of public figures, and do not write scripts that put words in a real person’s mouth to mislead anyone.

Your photo is used only to make your video. The prepared copy of the face is deleted when the video is done; the finished MP4 stays in your history.

Read the Terms of Service for the full rules.

AI talking photo FAQ

How do I make a photo talk?: Upload a portrait, type what it should say, pick a voice and press generate. You get an MP4 of the person in the photo speaking your script, with sound.
Do I need a separate text-to-speech tool?: No. The voice is generated from your script as part of the same job, already synced to the lips. You do not upload any audio.
How long can a talking photo be?: Up to 30 seconds, which is about 450 characters of English. For longer talks, generate several parts and join them in an editor.
How much does it cost?: 2 credits per second of speech, estimated from the script: 12 credits for the shortest clip and 60 for a 30-second one. Credits come with any NENE plan or credit pack.
Is the AI talking photo generator free?: No. Each video costs credits, shown before you generate. If a video fails, the credits are returned automatically.
Which languages can it speak?: English for now. Other languages will be added once they have been tested for natural speech and lip sync.
What kind of photo works best?: One person, facing the camera, face sharp and evenly lit, framed from the chest up, with nothing covering the mouth.
Can I use a photo of someone else?: Only an adult who agreed to be animated. Photos of children and public figures are not allowed, and neither are scripts meant to impersonate someone.
What is the difference between a talking photo and motion transfer?: A talking photo speaks your script with face and head motion. Motion Transfer AI makes your photo copy the body movements of a reference video, without a script.
Is there a watermark?: NENE does not add a watermark to the MP4 you download.
What happens to my photo?: It is used only to make your video. The prepared face copy is deleted when the video is finished; the MP4 stays in your history.