video model · by Pruna AI

P-Video Avatar AI Lip-Sync Video Generator

Turn one portrait and an audio track into a talking-head video with P-Video Avatar, Pruna AI's speech-driven avatar model. Upload a photo, add your recording or pick a Fliki voice, and get a lip-synced clip in 720p or 1080p. Steer expression and head motion with an optional text prompt, all inside Fliki. Compare it side by side with all our AI video models before you render.

Generated with P-Video Avatar

Talking-head clips generated with P-Video Avatar inside Fliki. One photo, one audio track, no edits.

Prompt

Energetic, friendly delivery. She starts with a slight lean toward the camera and raised eyebrows on 'This app', gives a small confident nod on 'three times a week', and finishes with a bright genuine smile and a tiny playful head tilt on 'Seriously, try it.' Handheld phone feel with very subtle natural camera sway. Keep the identity, wardrobe, background and lighting exactly as in the first frame. Natural blinks every few seconds, relaxed breathing visible in the shoulders, and small head movements that follow the rhythm of the speech. Mouth shapes stay precise and match every syllable of the audio. No sudden jumps, no warping of the face, no extra hands entering the frame.

Prompt

Warm, encouraging teacher energy. Gentle smile throughout, eyebrows lift slightly on 'very first', a small open-handed nod on 'Python function', and a confident forward nod on 'Let's go.' Static tripod camera with a very slow, subtle push-in. Keep the identity, wardrobe, background and lighting exactly as in the first frame. Natural blinks every few seconds, relaxed breathing visible in the shoulders, and small head movements that follow the rhythm of the speech. Mouth shapes stay precise and match every syllable of the audio. No sudden jumps, no warping of the face, no extra hands entering the frame.

Prompt

Polished broadcast delivery with friendly morning warmth. Steady, composed head position with small professional nods on 'sunshine' and 'seventy-five degrees', a brief warm smile at the end. Locked-off studio camera, no camera movement. Keep the identity, wardrobe, background and lighting exactly as in the first frame. Natural blinks every few seconds, relaxed breathing visible in the shoulders, and small head movements that follow the rhythm of the speech. Mouth shapes stay precise and match every syllable of the audio. No sudden jumps, no warping of the face, no extra hands entering the frame.

Prompt

Cheerful, bouncy cartoon performance. A playful little head bob on 'Beep boop!', eyes widen with excitement on 'I'm Bolt', and a friendly tilt of the head on 'how recycling works.' Antenna light gently pulses. Soft slow camera drift. Keep the identity, wardrobe, background and lighting exactly as in the first frame. Natural blinks every few seconds, relaxed breathing visible in the shoulders, and small head movements that follow the rhythm of the speech. Mouth shapes stay precise and match every syllable of the audio. No sudden jumps, no warping of the face, no extra hands entering the frame.

Prompt

Warm, helpful, clear delivery. Friendly smile on 'Bienvenidos', a small introduction nod on 'Soy Lucía', and a gentle reassuring head tilt on 'cómo abrir tu cuenta.' Static camera with a very slow push-in. Keep the identity, wardrobe, background and lighting exactly as in the first frame. Natural blinks every few seconds, relaxed breathing visible in the shoulders, and small head movements that follow the rhythm of the speech. Mouth shapes stay precise and match every syllable of the audio. No sudden jumps, no warping of the face, no extra hands entering the frame.

Prompt

Warm, confident agent delivery. Opens with a welcoming smile, a slight turn of the head toward the room on 'sunny three bedroom home' before returning to the lens, and an enthusiastic nod on 'just listed this week.' Slow gentle dolly in. Keep the identity, wardrobe, background and lighting exactly as in the first frame. Natural blinks every few seconds, relaxed breathing visible in the shoulders, and small head movements that follow the rhythm of the speech. Mouth shapes stay precise and match every syllable of the audio. No sudden jumps, no warping of the face, no extra hands entering the frame.

Prompt

Calm, reassuring, friendly delivery. A soft sympathetic nod on 'Thanks for reaching out', then a warm relieved smile and a small confident nod on 'Your refund is on its way today.' Static webcam-style framing, no camera movement. Keep the identity, wardrobe, background and lighting exactly as in the first frame. Natural blinks every few seconds, relaxed breathing visible in the shoulders, and small head movements that follow the rhythm of the speech. Mouth shapes stay precise and match every syllable of the audio. No sudden jumps, no warping of the face, no extra hands entering the frame.

100M+VIDEOS CREATED
14M+USERS WORLDWIDE
80+LANGUAGES SUPPORTED

Why creators choose P-Video Avatar

Photo plus audio in, talking video out

Give P-Video Avatar one portrait and one audio track. The model animates the face so the mouth moves with every word, with no video reference or recording session needed.

Lip sync driven by your audio

Timing, pauses, and mouth shapes come from the audio itself. Use your own recording for a personal message or any Fliki voice for a polished read.

Works on real and stylized faces

Pruna AI says the model handles photoreal people, illustrated characters, and stylized avatars alike, so mascots and 3D characters can speak too.

720p or 1080p output

Render drafts in 720p, then switch to 1080p for the final cut. Both resolutions are available on Fliki for P-Video Avatar.

Multilingual speech

Pruna AI lists English, Spanish, French, German, Italian, Portuguese, Japanese, Korean, and Hindi among supported languages. Localize one portrait into many markets.

Built for short presenter clips

On Fliki each P-Video Avatar generation is a short clip, ideal for hooks, UGC-style ads, intros, and reply videos. Stitch several together for longer segments.

Prompt-directed performance

Add a text prompt up to 2,048 characters to shape expression, head movement, gestures, and framing, so the delivery matches the tone of the script.

Output matches your photo

The output aspect ratio follows the input image. Upload a vertical portrait for Shorts and Reels or a landscape one for YouTube, with no awkward crop.

How it works

How to make a talking video with P-Video Avatar

Turning a portrait into a talking video with P-Video Avatar takes a few minutes inside Fliki. Follow these six steps.

Upload a front-facing portrait photo for P-Video Avatar lip-sync on Fliki
Step 1

Upload a portrait photo

Open the Fliki AI Playground in lip-sync mode and upload one clear, well-lit photo with the face frontal and fully visible. Head-and-shoulders shots sync more accurately than full-body shots.

Fliki model selector with P-Video Avatar chosen for lip-sync video
Step 2

Select P-Video Avatar as your model

Open the model selector and choose P-Video Avatar. Fliki sends your photo as the first frame and your audio as the speech track, with no extra setup.

Choose a vertical or landscape portrait to set the P-Video Avatar output aspect ratio
Step 3

Frame the photo for your aspect ratio

The output follows the shape of your photo. Use a vertical portrait for TikTok, Reels, and Shorts, or a landscape image for YouTube and web.

Add an audio file or generate speech with a Fliki voice for P-Video Avatar
Step 4

Add audio or pick a Fliki voice

Upload your own voice recording, or type a short script and generate it with any Fliki voice. The mouth, timing, and pauses follow the audio you provide.

Write an optional motion and expression prompt for P-Video Avatar on Fliki
Step 5

Direct the performance (optional)

Write a short prompt describing expression, head motion, gestures, and camera behavior, for example a warm smile, small nods, and a slow push-in. Up to 2,048 characters.

Pick 720p or 1080p and generate a talking video with P-Video Avatar on Fliki
Step 6

Pick resolution and generate

Choose 720p or 1080p and hit Generate. Preview the clip, download it, or drop it into a longer Fliki project.

AI MODEL GALLERY

Built on the best AI models - ready inside Fliki

Every leading video, voice, and image model - integrated, unified, and tuned for creators. Generate with the latest AI video, AI voice, and AI image models from OpenAI, Google, Kling, Bytedance, ElevenLabs, and more - all from one place.

P-Video Avatar FAQ

Frequently asked questions

Everything you need to know about generating with P-Video Avatar inside Fliki.

Still curious?

Try Fliki free in your browser, no credit card required.

Start free
P-Video Avatar · Paid plans

Make any portrait talk with P-Video Avatar.

Lip-synced talking-head video from one photo and one audio track, in 720p or 1080p. Available on paid Fliki plans.

Create your first talking video

Free forever plan · No credit card required · Cancel anytime