Live and real-time transcription
Daisy can show the transcript while you record. Everything runs on-device — whether live captions appear depends on whether your machine is fast enough, and Daisy measures that for you with a quick speed check. You can override the automatic choice any time.
What you see while recording
The live transcript separates speech by source: your system audio is labeled Them, and your own microphone lines appear alongside it unlabeled. Words appear as they're spoken and settle into clean text a moment later. Interim words look slightly faded and italic until they settle.
Captions appear in the recording view and, if you're using it, in the floating mini-window (which shows a tail of the most recent lines).
Before any speech arrives you'll see waiting for live transcript…. If
live captions are off for your machine, the recording view shows a card
titled Live captions are off on this machine, noting that recording
and transcription are still running and the full transcript is created
automatically when you stop.
How Daisy decides, per machine
Live captions run on-device with the bundled Whisper model — your audio never leaves your machine. Because keeping up with live speech takes real compute, Daisy runs a quick speed check on each machine (the setup wizard does it on first run) and remembers the verdict:
- Fast enough — live captions appear while you record.
- Not fast enough — no live overlay, and the recording view says so. The full transcript is still produced on-device the moment you stop.
The setting is per machine: your desktop can show live captions while your travel laptop skips them, from the same synced profile.
Overriding the automatic choice
Under Settings → Recordings → Live captions you can switch between Auto (follow the speed check), On, and Off. The Run speed check button re-measures this machine — useful after a hardware change or if you first ran the wizard on battery power.
You may see the same line twice — that's normal
During a call, the other person's words can appear twice — once on your own microphone line and once on the labeled Them line. This is expected and not a bug.
Daisy captures two independent audio streams:
- Your microphone — your voice. In the live view these lines carry no label.
- Them — the system audio coming out of your speakers (the other side of the call), labeled Them.
When you use speakers instead of headphones, your microphone picks up the other person's voice, so their words get transcribed on both sides. Daisy detects these mic-bleed duplicates by comparing the wording and drops one copy. In the live view it always drops the microphone copy and keeps the Them line. At finalize it goes a step further, comparing the two recordings acoustically to work out which copy is the echo — usually the mic copy, but it keeps whichever one is the true source.
Using headphones eliminates this at the source: the other side's audio never leaks back into your mic.
What changes when you stop
When you press ✦ Finish & summarize, Daisy doesn't start over. It runs echo cancellation, keeps or fills in the transcript, sorts out who said what on-device, removes the mic-bleed duplicates, and writes the summary. The live transcript is a preview — the saved transcript is the source of truth. See the FAQ for rough timings.