Turn a static image and a favourite song into a lifelike AI-generated singing video with realistic lip movements and natural facial expressions. Pick any portrait, drop in audio, and share it on TikTok and Reels.
Key featuresFeatures of Make Photo Sing
Pixel-perfect AI lip sync engine
Every syllable, breath, and beat in your favourite song lines up with the face. The AI lip sync enginesynchronises mouth shapes at the phoneme level, so each performance looks real, not robotic. Even held notes and runs land clean.
Smart face detection on any portrait
Drop in a selfie, a pet picture, a cartoon character, or a vintage portrait. The image-to-video pipeline reads facial features, locks the framing, and reconstructs subtle head movements so each animation stays consistent across long clips and close-up shots.
Bring your song or AI-generated voiceover
Drop in an MP3 or WAV at the full length of your track. Prefer a custom narration? Generate one in the AI voice generator and route it directly into the singing photo. Both options use the same advanced engine for consistent quality.
Expressive facial animation, not robotic
Choose a mood, and the AI face responds like a polished AI spokesperson. Subtle blinks, smiles, eyebrow lifts, and head tilts are added automatically, so a ballad feels gentle and a hype track feels fully charged. No frame-by-frame work or animation timeline to struggle with later.
Captions and vertical exports built in
Burn-in word-by-word captions with the subtitle generator and export 9:16 for Reels, 1:1 for feed, or 16:9 for YouTube. Safe text placement ensures your hook stays within the frame on every platform on the very first view, every time you post.
Use cases
Static selfies do not stop the scroll. Turn any photo into a singing image of the latest trend, post a few variants in a row, and let format and timing do the work for you across feeds.
A flat e-card feels like a task. Make your grandma sing happy birthday in her own voice, send the inside-joke version to your best friend, or use a custom song for a wedding toast that feels fun and engaging.
Pets are pure social fuel. Upload a clear photo of your dog, cat, or hamster, add a hilarious song so they can sing your favourite tunes, and enjoy I Love Happy Cats–style cuteness on every feed where you post.
Animate faces from illustrations, channel mascots, AI-generated cartoon characters, or your virtual identity without any rigging or motion capture. The animator gives every avatar a truly on-stage style performance, post after post, without blocking out any shooting time in your calendar.
Animate faces from illustrations, channel mascots, AI-generated cartoon characters, or your virtual identity without any rigging or motion capture. The animator gives every avatar a truly on-stage style performance, post after post, without needing to block any shooting time in your calendar.
Make historical figures sing their dates, animate book characters reading lyrics, or send a song through the AI video translator into 177+ languages so the same lesson is ready to share across multiple languages and classrooms.
How it worksHow the Make Photo Sing tool works
Make your photo sing in four simple steps. Animate photos with no shooting, no editing experience, and no plugins needed.
Choose a clear, front-facing image of a person, pet, or character. The AI will detect the facial features.
Upload an MP3 or WAV file, choose a song from your library, or paste lyrics for the AI to sing.
Choose an emotion to guide the performance, and select whether you want a vertical or square format for where you plan to post.
Render the AI photo output, preview the singing result, and download an MP4 that is ready to use on any platform.
It means turning a still image into a short video where the face performs an audio track you can even localise with AI dubbing. The AI will automatically detect facial features, map phonemes to mouth shapes, then add blinks and head tilts so the photo brings photos to life singing your chosen song.
Sign up online for a free HeyGen account, simply upload a clear front-facing portrait image, add an audio file, and click generate. The free plan covers short clips with a watermark, while a paid plan unlocks longer renders, HD output, and formats ready for AI video ads.
Clear, well-lit, front-facing headshot images work best. Avoid heavy occlusions, sunglasses, extreme side angles, and low resolution. Pet shots, cartoons, and illustrations are all fine as long as the face is visible and centred in the frame; for document sources, PDF to video is usually the better option.
Yes. Drop in any MP3 or WAV, including original tracks, covers, voice memos, instrumentals, or exports from the AI podcast generator. The AI reads vocals, beat, and tone, then maps the performance to match the rhythm. Confirm you hold rights to any commercial music before you post.
Yes. HeyGen supports multiple languages with natural pronunciation that matches the rhythm of each language. Run the vocal through the AI video translator into Spanish, Hindi, Mandarin, or any other language, and the lip-syncing will adjust automatically.
Yes. Open the clip in the AI video editor to trim, adjust expression intensity, swap audio, regenerate alternate takes, or add captions, music beds, and brand colours before exporting your final cut.
Yes. If the image has a recognisable face, the AI animates it. Dogs, cats, illustrated cartoon characters, anime portraits, and AI-generated avatars all work. Pet and cartoon singing is one of the most popular faceless video formats on the platform.
The tool animates one face at a time for the cleanest result. Crop the image to the person you want as the singer, run an AI face swap to change the star, or generate separate clips for each face and stitch them together later in post-production for a group scene.
Yes, you own the outputs. Please confirm that you hold the rights to any third-party music, photos, or voices you upload. HeyGen serves 85,000+ businesses, from solo creators like Anton Voroniuk reaching over 1M students to global brands.
Most short clips render in a few minutes. The AI-powered output is MP4, available in 9:16 vertical, 1:1 for feed, and 16:9 for YouTube. Captions can be burnt in for autoplay or kept separate for repurposing across channels, in the same way PPT to video reworks a slide deck.
Explore more AI-powered tools
Bring any photo to life with hyper-realistic voice and movement using Avatar IV.
Make any photo sing with AI in just a few minutes. Upload a picture, add a song, and share it on TikTok and Reels.
