About InstaScript

InstaScript is a free browser-based transcription tool. It turns video and audio files (MP4, MOV, WebM, MP3, WAV, M4A, OGG) into timestamped text, with .txt, .srt and .vtt downloads. The speech-transcription link input accepts public Instagram URLs. Separate YouTube tools read existing caption tracks from public YouTube links.

File processing runs in your browser; public-link tools also use server lookups. No app to install, no account to create, and no payment required.

How It Works

When you upload a video or audio file (or paste a public Instagram URL), InstaScript extracts the audio and runs it through Whisper, an open-source speech recognition model developed by OpenAI. The model runs locally in your browser using transformers.js, a library by Hugging Face that brings AI inference to the web.

This means your files never leave your device when using the upload feature. The transcription happens right on your computer or phone.

Features

Transcribe video files (MP4, MOV, WebM) and audio files (MP3, WAV, M4A, OGG) up to 100 MB
Paste a public Instagram URL and let the site fetch the video for you — this speech-transcription input supports Instagram links; separate YouTube tools read existing captions. For TikTok or Facebook, save the file and upload it
Timestamped output with start and end times for each segment
Copy transcript to clipboard with one click
Download as plain text (.txt) or subtitle formats (.srt and .vtt)
Works on desktop and mobile browsers
Limits: English audio only, and long recordings are processed in chunks of about 30 seconds

The Project

InstaScript is a personal project, published under the site name InstaScript — no individual is named or credited on this site. It is independently developed and maintained, and it is not affiliated with, endorsed by, or sponsored by Instagram, Meta, TikTok, ByteDance, Facebook, or OpenAI.

Questions about the site, or about your data, go to [email protected].

Tech Stack

Frontend: Vanilla HTML, CSS, and JavaScript (no framework)
AI Models: Whisper tiny for language detection and Whisper tiny English for transcription
Inference: @huggingface/transformers.js (runs in-browser via WASM)
Hosting: Cloudflare Pages with serverless Functions for URL fetching

Privacy

Local media files and subtitle edits stay in your browser. Public Instagram links use server-side media fetching; YouTube tools use server-side caption lookup, whose successful responses may be cached for one hour. See our Privacy Policy for details.