Local Transcript — On-Device Speech-to-Text for Chrome

How it works

Four steps, no setup

Open the side panel

Click the Local Transcript icon in your Chrome toolbar to open the panel on any tab.

Choose a save file

Pick a text file on your computer. The transcript will be written there in real time.

Press Start

The extension begins transcribing the tab's audio immediately. No waiting, no loading bar.

Press Stop — or walk away

Recording stops when you press Stop, when the video ends, or after a period of silence.

Features

Everything you need, nothing you don't

📴

Fully offline

The Whisper AI model ships inside the extension. No internet required after install.

💾

Saves as you go

Transcript is written to your file chunk by chunk. Close the browser mid-session — nothing is lost.

🌐

Works on any tab

Lectures, Teams calls, YouTube, podcasts, Zoom — any tab that plays audio can be transcribed.

⏱️

Optional timestamps

Each paragraph can be prefixed with its recording time so you can find moments in the original video.

⚡

GPU accelerated

Uses WebGPU when your hardware supports it, automatically falling back to WebAssembly.

🎯

Auto-stops

Recording ends automatically when the media finishes or after a silence — no babysitting required.

Privacy Policy

Last updated: October 2026

Overview

Local Transcript (Speech-to-Text) is a Chrome extension that transcribes audio from browser tabs using on-device speech recognition. This policy explains what data the extension accesses, how it is used, and what is never collected.

Data we do not collect

Local Transcript does not collect, transmit, or store any personal data. Specifically:

Audio captured from your browser tabs is processed locally and immediately discarded. It is never recorded to disk or sent anywhere.
Transcripts are written directly to a file you choose on your own computer. They are never uploaded to any server.
No analytics, crash reports, or usage telemetry are sent.
No account or login is required. The extension has no back-end service.

Permissions used

tabCapture — captures the audio output of the tab you choose to transcribe. Audio is processed on-device and never stored or transmitted.
offscreen — runs the speech recognition model in a background document so transcription continues without blocking the browser UI.
sidePanel — displays the live transcript in the Chrome side panel.
storage — saves your preferences (e.g. whether timestamps are enabled) locally in Chrome. No personal data is stored here.
activeTab — scopes audio capture to the specific tab you have opened the panel on.
scripting — injects a small listener to detect when media playback ends, so recording can stop automatically.
alarms — periodically checks whether the AI model should be unloaded from memory after a period of inactivity.

Third-party services

The extension uses the open-source Whisper model (whisper-base.en) by OpenAI, bundled inside the extension package. The model runs entirely on your device. No data is sent to OpenAI or any other third party.

Changes to this policy

If this policy changes, the updated version will be published at this URL. Significant changes will also be noted in the extension's Chrome Web Store changelog.

Contact

Questions about this policy? Email jerrychang0803@gmail.com.

Speech-to-text that stays on your device

Your audio never leaves your computer