---
title: "About audiogram.video: audio to video in your browser"
description: "Why audiogram.video exists, how it turns recordings into waveform videos on your own device, and what it promises: no audio upload, no watermark, no account."
url: https://audiogram.video/about
---

# About audiogram.video

Sound has a hard time on the social web. Instagram, TikTok, YouTube and LinkedIn all want a video, so a great podcast moment, a voice memo or a new song has nowhere to go until it has pictures. An audiogram is the simplest answer: the recording, a wave that moves with it, some artwork and the words on screen so people can follow with the sound off. (It is not the hearing-test chart that shares the name.)

This site exists to make that video quickly and well, without handing your recording to anyone. It is an independent, self-funded project with no investors and no advertising.

## What you can do here

* Open a recording of any length and pick the part you want, or let the studio find the liveliest stretch for you.
* Choose from seventeen templates, eight layouts and nine kinds of moving wave, then make them yours with your cover art, colours and fonts.
* Turn the speech into captions that light up word by word, fix any word by typing, and download them as SRT or VTT.
* Export square, vertical, portrait or widescreen MP4s at 1080p, ready for every platform, plus a still for the post’s cover.
* Save your show’s look as a style and use it on every episode.

## How it works

The page is a static web app. When you open a file, your browser decodes the sound with the WebCodecs API and the open-source mediabunny library, measures its loudness and frequencies frame by frame, and draws every frame of the video on a canvas. The same drawing code runs in the preview and the export, so what you see is what you get. The browser’s own video encoder then writes an H.264 MP4 with AAC sound. Formats the browser can’t read (WMA, AIFF, AMR) are converted on your device first by an open-source converter that loads only when it’s needed.

## The one thing that leaves your device

Writing out speech accurately takes a large speech-recognition model, too big to run well on a phone. So when you press **Make captions**, and only then, the sound of the clip you chose (never the whole recording, never your pictures) is compressed and sent through this site to our speech-recognition provider, which writes it out and keeps nothing. If you’d rather nothing left your device at all, type the words or open an SRT file instead. The[privacy page](https://audiogram.video/privacy) has the details.

## What we promise

* Your recording, cover art and photos are never uploaded to make the video.
* No watermark, no sign-up, no daily limit and no paid tier.
* No advertising. Analytics cookies can be switched off from Cookie settings in the footer, and the[privacy page](https://audiogram.video/privacy) lists every measurement tool.

## Who uses it

Podcasters sharing a clip of each new episode, interviewers and journalists pulling out a quote, musicians putting a song on YouTube with a visualiser, teachers and preachers sharing a short talk, and anyone with a voice note worth more than a chat bubble.

## Get in touch

Found a recording that won’t open, a template you wish existed or a bug? The[contact page](https://audiogram.video/contact) has the address.