About MeetingScribe

What this site is

MeetingScribe is a free tool that turns a recording into text without uploading the recording anywhere. You open the page, drop in an audio file, and the transcription happens inside your own browser, on your own computer. There is no account to create, no file to upload, no queue to wait in, and no copy of your conversation sitting on a server you do not control.

Alongside the tool, the site publishes practical guides on transcription, subtitles, recording law, and meeting notes. Those guides exist because the questions people ask before they transcribe something (is this legal, which format do I need, how do I get usable notes out of an hour of talking) are usually harder than the transcription itself.

Why it exists

Nearly every free transcription service works the same way: you upload your audio, their servers process it, and you get text back. That is a perfectly reasonable design for a podcast or a conference talk. It is a bad fit for a client intake call, a therapy session, an HR investigation, a source interview, a doctor's dictation, or a board discussion that has not been announced yet.

For that second category, the blocker is rarely price or accuracy. It is that the person cannot hand the recording to a third party, so they end up typing it out by hand or not transcribing it at all. Speech recognition models became small enough and browsers became capable enough that this no longer has to be the tradeoff. The model can come to the audio instead of the audio going to the model. MeetingScribe exists to make that the default rather than an enterprise feature.

Who publishes it

MeetingScribe is built, published, and maintained by SugarPinePM, a small independent software company. It is not a venture-funded startup and not a side project that might disappear next month. It is one product in a small portfolio, built and supported by the same people who wrote it.

Being small has real consequences, and they run in both directions. In your favor: there is no growth target that requires collecting your data, no sales team that needs a reason to email you, and no investor pressure to add an account system so usage can be measured. The product can stay narrow and stay private because nothing about the business model depends on it getting bigger.

Against: there is no 24 hour support desk, no uptime guarantee, and no enterprise contract. Fixes and new guides ship on a small team's schedule. If your organization requires a service level agreement or a signed data processing agreement, a larger commercial vendor is the right choice, and we would rather say so plainly than pretend otherwise.

Questions about the company, the site, or anything on it go to hello@meetingscribe.net, which reaches a person rather than a ticketing system.

How the site is funded

MeetingScribe is free and has no paid tier. The site is funded by advertising, shown through Google AdSense, and by nothing else. In particular:

• We do not sell, rent, or share your data. There is no data to sell: your audio and your transcript never reach us.
• We do not charge for the tool, gate features behind a signup, or limit how many files you transcribe.
• Ads are loaded only after you accept cookies in the consent banner. Decline, and the advertising script is never loaded at all.
• Ads are kept out of the transcript area and away from the download buttons, so an ad can never be mistaken for a control you meant to click.

We decided against a paid tier on purpose. The obvious thing to charge for would be a one click AI summary, which would mean sending your transcript to a server. That would sell back the exact thing the tool exists to protect. Instead the notes feature copies a ready made prompt to your clipboard so you can paste it into whichever AI chat you already use and trust. Read more in how it works.

What we will not do

These are commitments, not a roadmap we might quietly revise:

• We will not upload your audio. Not for processing, not for quality improvement, not for model training. This is the entire point of the product.
• We will not add an "email me this transcript" button. Sending a transcript through our servers would break the same promise more quietly.
• We will not require an account. Accounts exist to link activity to a person, and we do not want that link to exist.
• We will not claim certifications we do not hold. The tool is designed for confidential workflows. That is a design description, not a regulatory claim, and you remain responsible for your own compliance obligations.

The limits we are upfront about

Running a speech model in a browser tab has real constraints. We would rather you read them here than discover them mid-task:

• Length. Transcription is reliable up to roughly forty minutes of audio per job. Longer recordings can exhaust the memory a browser tab is allowed. Splitting a file into parts named in order works around this, and the parts are joined into one transcript.
• First run. The model downloads once, about 50 MB to 270 MB depending on which size you pick. After that it is cached in your browser and later runs start quickly.
• Hardware. A laptop or desktop with GPU acceleration in Chrome or Edge is substantially faster. Phones work for short clips but run on the processor and can be slow.
• Accuracy. This uses Whisper, which is strong on clear audio and degrades with crosstalk, heavy accents, and background noise, like every speech recognizer. The transcript is editable in the page before you download it, because no model gets everything right.
• Speaker labels. Speaker separation is most accurate with one to three voices. In a crowded meeting it will make mistakes, and you can rename or correct the labels yourself.

How the guides are written

The guides are written by hand, based on the tool as it actually behaves and on sources we checked at the time of writing. Where an article touches recording and consent law, it is general information rather than legal advice, and it says so, because the rules vary by state and by country and change over time. Where an article mentions other products, we describe the tradeoff honestly, including the cases where a cloud service is the better choice. If you find something out of date or wrong, we would like to know.

Get in touch

Corrections, bug reports, and feature requests are all welcome. Email hello@meetingscribe.net or use the contact page. If you want to understand the mechanics before you trust it with a recording, how it worksexplains what runs where, and shows you how to verify the privacy claim yourself in your browser's developer tools rather than taking our word for it.

Site details

• Published by: SugarPinePM
• Contact: hello@meetingscribe.net
• Funding: advertising only, through Google AdSense. No paid tier, no data sold.
• Content: written in house. Guides that touch law or third-party pricing are general information and are reviewed as they age.