If you're on a call and want Junior to capture it without inviting a bot, you can record straight in your browser. A live transcript, a running interview-guide checklist, and a notepad all appear while the conversation happens.
Start a browser recording
- Open the call, or create a new one, and choose Record in browser.
- Allow microphone access when your browser asks. Junior needs it to capture audio.
- Name your speakers if prompted, then start talking. A recording timer confirms Junior is capturing.
- When you're done, click End Call and confirm. Junior begins processing immediately.
Don't wear headphones while recording in your browser. Junior captures the call through your computer's microphone and speakers, so the other participants have to be coming through your speakers out loud. With headphones on, their audio goes to your ears instead of the mic and Junior records only your side of the conversation. Keep the call on your speakers so everyone is captured — or send the recording bot instead, which joins the meeting and captures every participant directly.
Browser recording captures your microphone live. If you already have a clean recording of the meeting, uploading the file gives a better transcript.
What you can do during a live call
- Live transcript: watch words appear as they're spoken, so you can confirm Junior is hearing everyone.
- Interview-guide checklist: watch your guide's questions tick off automatically as they're covered. See setting up a project.
- Notepad: jot notes to yourself. They're saved alongside the call.
- Live chat: ask Junior a question mid-call based on the live transcript.
- Speaker names: label who's speaking so the final transcript is attributed correctly.
You can also pop the recorder out into a small floating window so it stays visible while you use other tabs or apps during the call. Look for the pop-out control in the live view.
The person who starts the recording owns it. Teammates who open the same live call see a read-only view rather than a second recorder.
Why the live transcript and the final transcript differ
The live transcript is a fast, real-time draft produced while you talk, so you can follow along. After the call, Junior re-processes the full recording with more accurate models and then cleans it: fixing misheard words, correcting speaker labels, removing filler and false starts, and, for interviews, reshaping it into tidy question-and-answer blocks. The final version is the accurate one, and it'll read differently from what you saw live.
A few specific things people notice:
- Wording changes. The live draft prioritizes speed; the final transcript prioritizes accuracy.
- The final version is shorter and cleaner. Filler, repetition, and false starts are removed during cleaning.
- Speaker labels may be corrected once Junior has analyzed the whole recording.
Processing starts the moment you end the call, and the takeaways and summary appear a few minutes later. See the call page & processing states.
If your connection drops mid-call
A dropped connection doesn't lose your recording. If your internet is interrupted, you'll see a message that the connection was interrupted and is reconnecting, but Junior keeps recording locally on your device and uploads it once you're back online. The live transcript may pause; the final transcript will still be complete.
You may also see a "We can't hear you, check your microphone" warning if Junior detects silence — a cue that your mic may be muted or the wrong input is selected.
Questions
Can I wear headphones while recording in my browser?
No. Browser recording captures the conversation through your computer's microphone and speakers, so the other participants have to be audible on your speakers. With headphones on, their audio goes to your ears instead of the mic, and Junior records only your side. Keep the call on your speakers, or send the recording bot — it joins the meeting and captures everyone's audio directly, so headphones are fine.
Why does the final transcript read differently from what I saw live?
The live transcript is a fast draft made while you talk. Afterwards Junior re-processes the whole recording with more accurate models and cleans it — fixing misheard words, correcting speaker labels, and removing filler — so the final version is shorter, tidier, and more accurate than the live draft. Expect it to read differently.