Recording
Use the Voice Capture page to record, transcribe, and create documents from your voice.
The Voice Capture page is a dedicated recording interface. Open it from the sidebar, choose your settings, and start talking. Everything you say is transcribed in real time, and when you're done, a polished document is created automatically.
Starting a recording
Open Voice Capture
Navigate to Voice Capture in the sidebar (look for the microphone icon).
Choose your environment
Select an environment preset from the pill buttons below the record button: Quiet, Office (default), or Noisy. This adjusts how aggressively background noise is filtered. See environment presets for details.
Check your language
Your language is shown in a small dropdown in the top-right corner. It defaults to English. Change it if needed — see supported languages for the full list.
Hit record
Click the large record button — the red circle in the center of the page. Recording begins immediately.
Allow microphone access
The first time you record, your browser will ask for permission to use your microphone. Click Allow. You only need to do this once.
During a recording
Once recording starts, you'll see:
- A status badge at the top showing "Recording" with a pulsing red indicator, elapsed time, and your selected language
- A live transcript panel below that updates as you speak — words appear in real time
- Auto-scrolling that keeps the latest text visible as you talk
Speak naturally at a comfortable pace. The transcription handles conversational speech well — no need to slow down or over-enunciate.
Multi-speaker recordings
When multiple people are speaking — in a meeting, interview, or group discussion — Voice Capture automatically detects each speaker. The live transcript shows speaker changes inline with color-coded labels:
- Each speaker gets a distinct color (blue, green, purple, orange) and a label like "Speaker 1," "Speaker 2"
- Labels only appear when the speaker changes, keeping the transcript clean
- If only one person is speaking, no speaker labels are shown
Pause and resume
Need a moment? Click Pause to temporarily stop recording. The badge switches to "Paused" and the timer freezes. Nothing is lost — click Resume to continue right where you left off.
This is useful when you need to gather your thoughts, answer a quick question, or take a break in a longer recording session.
Stopping and saving
When you're finished, click Stop & Save. Here's what happens next:
- The status changes to "Creating your document..."
- Your audio is uploaded to Opisense
- The AI cleans up your transcript — fixing grammar, removing filler words, and organizing the text into readable paragraphs
- If multiple speakers were detected, the AI produces structured meeting notes with a summary, speaker-attributed discussion, and action items
- You're automatically redirected to the finished document in Your Content
The whole process takes a few seconds for short recordings, and up to a minute for longer ones.
Recordings shorter than 2 seconds are discarded automatically. If you accidentally tap record, just stop quickly and nothing will be saved.
Recording on mobile? The mobile app has a dedicated Voice Capture screen with the same real-time transcription. See Voice Capture on mobile for details.