Most transcription apps send your audio to their servers. Scrieb doesn't. Recording, transcription, speaker labels, summaries and AI chat all run on your own computer - Mac or Windows. No cloud. No uploads. Built for lawyers, therapists, researchers and anyone whose conversations can't be shared. Turn meetings, calls and voice notes into clean text that shows who said what, ask the transcript questions with a local AI chat, then export to DOCX or PDF - every word still on your device.

I'm Alex, and I built Scrieb because I kept running into the same problem: lawyers, therapists, and consultants who desperately needed transcription but couldn't use any of the popular tools - because their clients' conversations can't be uploaded to someone else's server. So I built one that doesn't. Scrieb records and transcribes entirely on your Mac & Windows. Nothing goes to the cloud, nothing touches my servers, nothing leaves your machine. If you work with sensitive conversations and have been avoiding transcription tools for that reason - this is for you. Curious what workflows you've been handling manually because of this

Just released today a fresh update 1.0.0:

• No more echo in calls: when the other person plays through your speakers, your microphone hears them a second time. Scrieb now removes that echo when the recording is finished. Recordings made with headphones stay exactly as they are.

• The spoken language is detected from the speech itself. A recording that starts with silence or music is no longer transcribed in the wrong language.

• Silent passages are skipped, so no invented text appears in quiet parts.

• A transcript that doesn't match the speech is marked "Check transcript", and you can transcribe it again with one click.

• Choose the spoken language yourself: for all recordings in Settings - Transcription, or for a single recording when transcribing again.

• Export to Word (.docx), with real headings, lists and speaker names, ready for Word, Pages and Google Docs.