
WhoScribe
An AI product that turns audio and video into speaker-labelled text in minutes
For WhoScribe we built the product on three legs: the web app, the iOS and Android app, and a Chrome extension that adds subtitles to any video in one click. All three share the same account, processing pipeline and subscription rights.
A meeting, lecture, interview or podcast: the user uploads a recording or pastes a link and, minutes later, reads who said what with timestamps, translates it and exports it.
Audio in. Clear, speaker-labelled text out.
Web App
Upload, transcribe, download
Built on Laravel, the app accepts files up to 5 GB through chunked upload; audio is extracted and split into segments of up to 10 minutes, processed in parallel on a queue and stitched back together with time offsets. Recordings that last hours become one clean transcript.
There are two modes: fast, and most-accurate with speaker separation. Text exports to DOCX, PDF, TXT, SRT and VTT and can be translated; an AI chat about the recording pulls out summaries, decisions and next steps.



Mobile App
Your recorder now writes
Built with React Native (Expo), the app shares the web account, processing pipeline, quota and purchase rights one to one. Users can record, import an audio or video file from the phone, or paste a link; the text can be searched, translated and chatted with.
The interface runs in 15 languages, including right-to-left Arabic. Users can start as a guest without signing up; store purchases are verified on the server.



Chrome Extension
Transcribe the video you are watching in one click
The AI Video Transcribe extension adds a small button to videos on the web. One click transcribes the audio; subtitles play in the user’s language inside the video’s own player, including full screen. The text can be read, searched or saved to the account.



Website and Growth Pages
We designed whoscribe.ai as a site that works like the product itself: visitors can transcribe their first file from the homepage without signing up. Searchable pages for sources such as YouTube, WhatsApp voice messages and Google Drive and for dozens of languages, plus pricing, mobile app and extension pages, run on the same design system in 15 languages.























