Private audio workspace

Find the exact moment it was said.

Hours of talks, calls, and interviews in one library, every word searchable to the second. The transcript is written by the mere.run app on your own computer, not by a hosted AI service.

Join waitlist Invited? Sign in

Invite-only while we scale. The waitlist lives on mere.world.

Live sample

Field recording, fisheries policy roundtable

1h 20m Ready

Try

2:41 Speaker B
The fisheries policy timeline moved after the June call, especially around cooperative reporting and permit renewals.
18:14 Speaker B
Search has to bring back the policy context even when someone only remembers the phrase cooperative reporting.

2 moments matched across 1h 20m. Nothing was uploaded — this search ran in your browser.

Two modes

You decide where the audio lives, before the first upload.

Private workspace

browser

Browser mode stores audio in your private workspace. mere.run on your computer collects it over an authenticated channel, transcribes it there, and sends only the transcript back.

Upload from any browser you are signed into
Play the source audio back beside its transcript
Delete the audio and keep the transcript, or delete both

Fully local

command line

Fully local mode keeps both the audio and library on your machine. Nothing is uploaded, and search runs against a database file you own.

mere-media import file ~/Audio/roundtable.m4a --store local mere-media search "cooperative reporting" --store local See fully local setup

Both modes need the mere.run app open on your own computer, on macOS or Linux. That is where transcription actually happens.

Private by default

Your transcripts stay yours.

Transcription runs on hardware you control, and the whole library exports to a single file, so leaving is one command.

Join waitlist on mere.world