Five ways to hand over a file
The Transcribe tab is in Settings. Drop a recording on it, or click Choose File, then press Transcribe. It takes audio and video that macOS can play, MP3, M4A, WAV, AIFF, FLAC, MP4 and MOV among them. A video is read for its sound only.
The first transcript uses the voice model you already dictate with, so there is nothing more to download. And you do not have to open Settings first:
- Drop the file on the Flit icon in the menu bar
- Right-click it in Finder and choose Transcribe with Flit
- Right-click it, choose Open With, then Flit
- Right-click the Flit icon and choose Transcribe File…, to set the model or formatting before it starts
While it runs
The text appears as each piece finishes, with how far into the recording Flit is and about how long is left. The file is read five minutes at a time, so there is no length limit and a two-hour recording uses about as much memory as a short one. In Flit’s own tests with Cohere Transcribe, two hours took about three and a half minutes at Clean and about fifteen at Polished.
You can close Settings and keep working. The menu bar icon wears a ring that fills as the job goes, and a notification says when the transcript is ready. Several files wait in line and run one after another. Stop ends a job and keeps what it has.
Read, copy and export
A finished transcript shows its length, word count, voice model and formatting level. A new paragraph starts where the speaker paused for two seconds or more. A switch puts Raw beside the formatted text: what the voice model heard, with every hesitation left in. Copy takes whichever side is showing.
The bin on a row deletes that transcript from Flit. The text file in your folder is yours, and it is left where it is.
- Export as Text, Markdown, SRT or VTT
- Subtitle files carry timestamps, one line per sentence
- Every transcript is also saved as a text file, in Documents → Flit Transcripts unless you choose another folder
- Past transcripts stay in the tab, with a search field once there are more than five
Its own model and formatting
Recordings have their own settings, separate from dictation, so you can dictate with a fast model and transcribe with an accurate one. Verbatim keeps every word. Clean removes hesitations and writes times and numbers properly. Polished runs the formatting model over each paragraph so the recording reads as prose.
What it does not do yet
A transcript is one flow of text. Flit does not say who said what, and it does not record a meeting as it happens: you give it a recording you already have. Speaker labels and meeting recording are both planned, and both will run on the Mac.
How to use it
- 1
Open the Transcribe tab
Right-click the Flit icon in the menu bar, choose Settings…, then Transcribe in the sidebar.
- 2
Add a recording
Drop it on the tab, or click Choose File and pick one.
- 3
Press Transcribe
The file is shown with its length, and the text appears as each piece finishes.
- 4
Copy or export
Copy puts the text on the clipboard. Export saves Text, Markdown, SRT or VTT.
The details
- Formats
- MP3, M4A, WAV, AIFF, FLAC, MP4, MOV
- Length limit
- None; the file is read five minutes at a time
- Speed
- Two hours in about 3.5 minutes at Clean (Cohere Transcribe, Flit’s tests)
- Exports
- Text, Markdown, SRT, VTT
- Saved to
- Documents → Flit Transcripts, or a folder you choose
- Speaker labels
- Not yet
- Languages
- Those your voice model covers
- Since
- Flit 1.5
Questions
Is my recording uploaded anywhere?
No. It is read on your Mac by the same voice models that dictate, and it works with Wi-Fi off.
How long can a recording be?
There is no limit. The file is read five minutes at a time, so a long recording does not fill memory.
Can Flit make subtitles for a video?
Yes. Export SRT or VTT and you get timestamps, with one line per sentence.
Does it label speakers?
Not yet. Speaker labels and recording a meeting as it happens are planned.
Which languages can a recording be in?
Whichever your voice model covers: Parakeet Ultra covers 25 European languages, Cohere Transcribe 14, Qwen3-ASR 30, Whisper 99 and Omnilingual more than 1,600.
The same feature on the app's own site: flit.fyi/changelog/1.5