By the numbers
0
Cloud calls during transcription
2
Audio channels captured simultaneously
$4.99
Pro unlock — once, forever
5s
Audio chunk window — volatile, never written to disk
How it works
Browser Tab Audio & Your Microphone
two independent channels
↓
Local Web Audio Graph
silence-gated · processed in browser memory
↓
Whisper AI — on your device
WebAssembly · no network · no API key
↓
Phonetic Brain Corrections
applied before you see the output
↓
Live Transcript · Studio View
speaker-labeled · timestamped · stored locally
↓
Export · .txt
your file · your machine · your call
01
Open your meeting in a browser tab
TrainScription captures audio from the browser's own audio pipeline — not from the meeting's participant list. Works with Google Meet, Teams (web), Zoom (web), Webex, Discord, and any other browser-based call. No third-party participant joins the call. No external transcription service relays your audio through cloud infrastructure. The meeting is simply a browser tab.
02
Click Begin Capture in the popup
The extension captures two audio streams: your microphone and the tab audio (the meeting). Both feed into a local Web Audio graph. A silence gate drops empty chunks before Whisper ever sees them, preventing hallucination. The AI runs entirely in your browser via WebAssembly — transcription occurs entirely on-device after the AI model has been downloaded. No network requests during transcription, no server, no API key.
03
Watch the transcript appear in real time
Every 5 seconds, a new chunk of speech is transcribed and appended — speaker-labeled with your configured channel names and a precise timestamp. Open the full Studio view to see the live transcript, or just leave the popup open in the background. Close the popup at any time — transcription continues.
04
Highlight misfires to train the Brain
See a word the AI got wrong? Highlight it in the transcript. A popover appears instantly — assign it to the correct spelling in one click. That phonetic mapping is now permanent. Every future transcript corrects that word automatically, silently, before you ever see it. The transcript stays visible after you end a session — no need to dig through Recovery. Finish your corrections, then clear when you're done.
05
Sessions are always recoverable
Long sessions are automatically segmented into sealed chunks. All segments from a single Begin→End session are grouped together in Recovery. Export any segment or the full session as a clean .txt file. Review any past segment in the live transcript view to catch and correct misfires after the fact.
Built for
Legal, Finance & Consulting
Professionals handling sensitive conversations where routing audio through third-party cloud services is simply not an option.
Researchers & Interviewers
Capture browser-based interviews without routing audio through third-party cloud transcription systems. Locally stored, fully exportable.
Heavy Meeting Users
Stop paying monthly transcription subscriptions. One payment covers unlimited sessions — no per-seat, no per-minute billing ever again.
The differentiator
The problem every AI transcript has
proper names
product names
case names
medication names
campaign terms
company jargon
character names
niche terminology
AI transcription tools consistently fail on specialized vocabulary — in predictable, repeatable patterns.
The Phonetic Brain is the permanent fix. Train it once on a word; it corrects that word on every future transcript automatically, before you ever see the output.
The Phonetic Brain
Whisper — the AI model at the core of TrainScription — is remarkable with general language and terrible with your language. Proper names, product names, case names, medication names, character names, niche jargon: all mangled, all the time, in predictable patterns.
The Phonetic Brain is a local correction layer you build yourself. Train it once on a word; it fixes that word on every future transcript, automatically, before you ever see the output. It compounds — the more you use it, the more accurate your transcripts become.
01 · FASTEST
Highlight in the live transcript
Select any misheard word during a live session or while reviewing a past segment. A popover appears — assign it to the correct spelling in under three seconds. No separate screen.
02 · MANUAL
Type it in the Brain panel
Already know how Whisper is mangling a term? Use the Manual Add form — correct spelling plus the variants Whisper produces. Instant.
03 · IMPORT
Restore from a .train.json backup
Share a specialized vocabulary set across installs, or bring your Brain to a new machine. The file is plain JSON — human-readable and editable outside the extension.
04 · ON THE GO
Train from your phone — before the meeting starts
Know a tricky word is coming? The free
BrainTrainer companion runs on Android and iOS. Speak the word a few times, pick the mishears, export a
.train.json, and import it into TrainScription Pro. Your Brain knows the word before the first transcript line appears.
05 · MANAGE
Remove variants or whole terms anytime
Made a wrong assignment? Click the Brain panel, expand any term, and remove individual variants — or delete the whole term — with a single tap. No JSON editing required.
Train before the meeting. Refine during the meeting. Keep improving after the meeting.
Everything included
No Meeting Bot Required
TrainScription works directly from browser-tab audio. No third-party participant joins the call. No external transcription service ever relays your audio through cloud infrastructure. The meeting is simply a browser tab.
Dual-Channel, Speaker-Labeled
Your microphone and the meeting audio captured separately. Speaker attribution by audio energy. Each line labeled with your configured channel names and a precise timestamp.
Phonetic Brain
Permanently fixes the words AI always gets wrong. Train by highlighting in the transcript. Corrections apply to every future session automatically — and every word the Brain corrects is highlighted live in the transcript so you can see the engine working in real time.
Channel Presets
Six built-in label presets for common use cases: Default, Corporate, Legal, Interview, Medical, RPG. Fully customizable per session from the popup.
Pause Controls
Pause your mic channel, the meeting audio, or everything — without ending the session. Critical for when you mute yourself in the call but don't want your in-room conversation transcribed.
Export to .txt
Download any session or individual segment as a clean .txt file. Pro users get one-click assembled export of the full session — all segments merged, formatted, ready to use.
Session Recovery
Sessions are automatically segmented and stored. Review any past segment in the live transcript view to find and correct misfires. Download individual segments as .txt anytime — or use Pro's one-click assembled session export.
Brain Management
Made a wrong assignment? Expand any Brain term and remove individual variants with a single click — or delete the whole term. No config files, no JSON editing. Your vocabulary stays exactly right.
Pricing
Free
$0
No credit card. No account required.
- 15-minute session cap
- 3 Recovery sessions stored
- 3 Phonetic Brain slots
- Real-time dual-channel transcript
- Pause controls
- Export capture to .txt after each session
Pro
$4.99 once
One payment. No subscription. Works across devices.
- Unlimited session length
- Unlimited Recovery sessions
- Unlimited Brain slots
- One-click assembled session export
- Import full Brain vocabulary files
- Everything in Free, forever
- Cross-device via email login
How it compares
|
TrainScription |
Otter.ai |
Fireflies |
Rev |
| No bot joins the call |
✓ Never |
✗ |
✗ |
✗ |
| Audio stays on your device |
✓ Always |
✗ |
✗ |
✗ |
| Works fully offline |
✓ |
✗ |
✗ |
✗ |
| Fixes jargon / names |
✓ Phonetic Brain |
Paid only |
✗ |
✗ |
| Pre-session vocabulary training |
✓ BrainTrainer |
✗ |
✗ |
✗ |
| Corrections visible in transcript |
✓ Live highlights |
✗ |
✗ |
✗ |
| Price |
Free / $4.99 once |
$16.99/mo |
$10/mo |
$9.99/mo |
| Works in any browser tab |
✓ |
Partial |
Partial |
✗ |
Why it exists
Most transcription tools route sensitive conversations through external cloud infrastructure, inject third-party participants into your calls, and charge you monthly for the privilege. The costs compound — financially, architecturally, and in trust.
TrainScription was built around a different idea: transcription should happen locally, under your control, on your own machine — regardless of how long the session runs or how sensitive the content is. The constraint of doing it entirely on-device isn't a limitation. It's the whole point.
Under the hood
TrainScription is a Manifest V3 Chrome Extension running
OpenAI Whisper via WebAssembly locally in your browser.
The AI model is downloaded once from the model provider and cached —
transcription itself occurs entirely on-device after that initial download.
TrainScription does not operate server-side infrastructure and audio is processed locally on your device and is not uploaded for cloud transcription.
Audio is processed in 5-second volatile chunks that exist only in RAM.
They are never written to disk. Audio is not uploaded for cloud processing.
The extension's data policy is architecture, not a promise —
the extension contains no code to route your audio externally.
The only external connections in this extension: a one-time AI model download
from the model provider, and ExtPay license verification for Pro users.
Both are unrelated to your content — your audio, transcripts, and Brain vocabulary
are never involved in either connection. TrainScription does not execute
remote code or load external scripts that alter extension functionality
after installation.
The Phonetic Brain correction function runs synchronously on every transcript
line before it's written to storage. Your vocabulary corrections apply
before you see the output.
Privacy & data
✓
Audio is processed locally using WebAssembly-based Whisper inference running in your browser. Audio chunks exist temporarily in RAM during transcription and are discarded immediately after processing. They are never written to disk.
✓
Audio is not transmitted to external servers. TrainScription does not operate cloud infrastructure and does not upload your audio for remote processing. Audio is processed on your device and is not routed externally.
✓
Transcripts are stored locally in your browser's extension storage on your machine. They are never synchronized, indexed, or accessible outside your device unless you export them yourself.
✓
Brain vocabulary corrections are stored locally in browser extension storage. Exports are user-controlled .train.json files that live wherever you save them.
✓
No account is required for the free version. No email, no registration, no profile.
✓
No analytics, no tracking, no behavioral data collection. TrainScription does not use tracking pixels, session recorders, or any analytics tooling.
✓
The extension works fully offline after the AI model has been downloaded. No internet connection is required for transcription.
External connections, fully disclosed: The AI model is downloaded once from the model provider (not from TrainScription) and cached locally. Pro license verification is handled by ExtPay — payment processing only. Your audio, transcripts, and Brain data are not involved in either connection.
TrainScription is intended for lawful, disclosed, and consent-compliant use. Users are responsible for complying with applicable local laws, workplace policies, and meeting platform terms regarding transcription and recording.
Why TrainScription requests permissions
Microphone
Used to capture your local microphone channel during transcription sessions. Only active when you initiate a capture session.
Tab Audio
Used to capture the meeting audio from the active browser tab during transcription. Only active when you initiate a capture session.
Storage
Used to save transcripts, Brain vocabulary corrections, and your settings locally on your device. Nothing is stored outside your browser's local extension storage.
Downloads
Used only when you explicitly export a transcript or Brain backup as a file. Not used at any other time.
TrainScription does not use permissions for advertising, tracking, behavioral profiling, or any purpose beyond the transcription workflow described above.