Audio to Video: Turn Any Audio into a Video with Captions

Turn a song, a WhatsApp voice note or a podcast clip into a video you can post: a background or your own picture, a waveform that moves with the sound, and the words on screen. It's made on your device, free, with no watermark.

Your audio never leaves your device. How?

In Accurate mode only the audio is sent to our server to be transcribed, then deleted. Choose "On this device" in the settings and nothing is uploaded.

Select or drop an audio file

A song, voice note, podcast or recording: MP3, M4A, WAV, OGG, OPUS (WhatsApp), FLAC, AMR. It stays on your device.

Click, tap, or drop a file

How to turn audio into a video

Choose the audio

Drop or select an MP3, M4A, WAV, OGG, OPUS (WhatsApp voice note), FLAC or AMR file.

Let AI write the words

The words are written with their timing, usually in under a minute. Click any line to fix a word.

Pick the look

Choose the shape (9:16 for Reels and Shorts, 1:1 or 16:9 for YouTube), a background or your own picture, how the waveform moves, and a caption style.

Create and download

The video is made on your device as an MP4, ready to post anywhere that takes videos.

What people make with it

Instagram, TikTok, YouTube Shorts and WhatsApp Status only take videos, so a song, a voice note or a podcast clip needs a picture before it can be posted. Podcasters share a minute of an episode with the words on screen (an audiogram), singers post a cover with the lyrics as they're sung, poets and speakers post a recitation, and people turn a voice note into something they can share.

YouTube also takes only video, so an MP3 has to become an MP4 first. Here it gets a picture, a moving waveform and the words, instead of a frozen image.

How the video is made

The words are written by the same speech recognition as Auto Subtitles: in Accurate mode the sound is sent to our server, transcribed and deleted; choose "On this device" and nothing leaves your device. Each word keeps its time, so the captions follow the voice.

The video itself is drawn and encoded on your device: every frame is drawn on a canvas (the background or your picture, the waveform at that moment of the sound, the words) and encoded as H.264 by your device's own video encoder. The sound goes in as it is when it's AAC or MP3, or is encoded to AAC. Your picture never leaves your device.

Features

Words in time with the voice

AI writes the words with their timing; each word lights up as it's said or sung.

Every shape

9:16 for Reels, Shorts, TikTok and Status; 1:1 for posts; 16:9 for YouTube.

Your picture or a colour

A cover, a photo or your logo, shown in the middle with the waveform under it, or a colour gradient.

Pro looks

19 more caption styles (Karaoke, Neon, Beast…), key words that pop with an emoji, and a circle waveform around your picture. Try any of them as a free 15-second sample of your own video.

Frequently asked questions

Is my audio uploaded?

Only to write the words, in Accurate mode: the sound is sent to our server, transcribed and deleted. Choose "On this device" and nothing leaves your device. The video, and your picture, are made and kept on your device.

Is there a watermark?

No. Every video is free of watermarks, with 8 free caption styles, the bars and wave waveforms and any background or picture.

Does it work for songs?

Yes: the words come out as they're sung, and each one lights up in time. Fast rap, heavy music over the voice or mumbled words can leave mistakes, so check the lines and click any word to fix it before making the video.

Which files work, and what comes out?

MP3, M4A, WAV, OGG, OPUS (WhatsApp voice notes), FLAC and AMR. The video is an MP4 (H.264 picture, AAC sound), which Instagram, YouTube, TikTok, Facebook and WhatsApp all take.

How long can the audio be?

Up to 15 minutes for the words in Accurate mode (longer is done on your device). For Reels, Shorts and TikTok, cut the part you want first with the Audio Cutter: a short clip is quicker to make and to watch.

Can I use my own picture?

Yes. Pick any photo, cover or logo: it's shown in the middle with the waveform under it, and blurred behind it so the words stand out. It stays on your device.