Accurate transcription, captured at the source
Nod transcribes the real call audio your Mac is playing rather than reading on-screen captions, so it isn't limited to a single browser tab and handles multiple languages. The result is a verbatim transcript alongside a clean summary of topics, decisions, and action items.
- Full transcript plus a structured summary
- Captures system audio at the OS level — not caption scraping
- Works across Zoom, Meet, Teams, and calls on speaker
- Searchable across every meeting with 'Ask Nod'
Transcription without a stored recording
Most transcription keeps an audio file by design. Nod doesn't. Capture is local on your Mac; transcription isn't. Each audio segment is sent through a European proxy to a hosted transcription service and discarded the instant the text returns, under Zero Data Retention and with no model training. No audio is stored at either end. Only the transcript and summary are saved, encrypted at rest in the EU (AWS eu-west-1). See the security page.
Related
See private, GDPR-compliant meeting transcription for the privacy detail, or Google Meet transcription without recording for a per-platform walkthrough.
Frequently asked questions
Can you transcribe a meeting without recording it?
Yes. Nod transcribes your Mac's audio in memory and discards it right after, so a transcript exists but no recording is ever saved, on your Mac or in any cloud.How accurate is Nod's transcription?
Nod transcribes the real call audio rather than on-screen captions, and supports multiple languages. Because it captures system audio at the OS level, it isn't limited to one browser tab the way caption-reading tools are.Is meeting transcription still subject to consent laws?
Yes. Keeping only the transcript is generally treated as recording the conversation, so get consent where one-party or all-party rules require it — see our consent guide.