Record your lecture.
Get study notes.
Drop a lecture recording — even a full 2-hour class — and get structured notes with chapters, summaries and timestamps. Free for a whole semester of lectures, because it runs in your browser.
Free plan today: WAV files of any length; MP3/M4A up to 1 hour per file for now. Phone recordings are usually M4A — for a class longer than an hour, export to WAV or split the file, or use Pro cloud mode.
From 2 hours of audio to something you can revise from
A raw transcript of a 2-hour class is 15,000+ words of wall-to-wall text — technically accurate, practically unreadable. Record to Notes structures it while it transcribes: chapters are detected at real topic changes, each chapter gets its own summary, and every line keeps its timestamp so you can jump back to the exact moment the professor explained it.
Record or upload
Phone recording, Zoom audio, dictaphone file — drop it in as-is. Video files work too; the audio track is used.
Notes appear live
Text streams in while it transcribes, so you can start reviewing the opening topics before the file is done. A 2-hour lecture takes about 10 minutes locally.
Study from it
Chapters with per-chapter summaries, exported as TXT, SRT or Markdown — drop the Markdown straight into Notion or Obsidian. Pro adds glossary, quiz and Anki flashcards.
Why students use this over the usual free tiers
Most transcription tools run your audio on their servers, pay per minute for it, and claw the cost back with free-tier caps. A semester of lectures blows through those caps in the first week. Record to Notes transcribes on your own device instead, so there is nothing to cap.
| Tool | Free plan limits | A 15-week semester of 2h lectures |
|---|---|---|
| Record to Notes | No length cap, no file cap, no signup (WAV; MP3/M4A ≤1h per file) | All of it, free |
| Otter.ai | ~300 min/month, 30 min per conversation | ~2 lectures' worth per month, split into 30-min chunks |
| TurboScribe | 3 files per day, 30 min each | 90 min per day maximum |
| Notta | Monthly minute quota with short per-file caps | A fraction of one week |
Competitor limits as of September 2026 — plans change, check their pricing pages. The structural difference doesn't: server-side tools have a per-minute cost to recover; on-device transcription doesn't.
What it handles, concretely
- Formats: WAV at any length. MP3 and M4A up to 1 hour per file for now (that cap is on compressed-audio decoding, and it's planned to go). Video files are accepted — the audio track is extracted.
- Speed: about 10 minutes for a 2-hour file on a recent laptop with GPU acceleration (WebGPU). On machines without it, transcription falls back to a slower universal mode — it still finishes, it just takes longer.
- Model choice: three local models — Tiny (~40MB, fastest), Base (~80MB, balanced), Small (~250MB, most accurate). Downloaded once, then cached by your browser.
- Privacy: transcription itself makes zero network calls. After the model loads you can watch the network tab, or switch off Wi-Fi mid-lecture — it keeps going. The recording never leaves your computer on the free plan.
- Output: full transcript with timestamps, auto-detected chapters, a summary per chapter. Export TXT, SRT or Markdown.
Pro — built for exam season
- Study Packs: structured notes, a glossary of the technical terms your professor actually used, a self-test quiz, and one-click Anki flashcard export
- Cloud speed: a 2-hour lecture transcribed in about a minute (vs ~10 minutes locally) — useful the night before the exam
- Highest accuracy in 90+ languages — echoing lecture halls, fast talkers, technical vocabulary
- Batch queue: drop the whole semester's folder and let it run
FAQ
Is it really free for long lectures?
Yes. Transcription runs on your own device, so there's no per-minute cloud bill to pass on to you. Unlimited lectures, unlimited length.
Can it transcribe a 2 hour lecture?
Yes — we test on 2-hour files routinely. Locally it takes about 10 minutes; with Pro cloud mode about a minute. One caveat: 2-hour files need to be WAV on the free plan — MP3/M4A is capped at 1 hour per file for now.
What file formats can I upload?
WAV at any length; MP3 and M4A up to 1 hour per file. Video files work too — the audio track is used. Phone voice-memo apps usually record M4A, so for classes over an hour, export to WAV, split the recording, or use Pro.
Is my recording uploaded anywhere?
Not on the free plan. The audio is processed entirely in your browser and never leaves your computer — transcription makes zero network calls once the model has loaded.
How accurate is it in a noisy lecture hall?
The local Small model handles clear classroom audio well. For echoing halls, heavy accents or dense technical vocabulary, Pro's cloud model is noticeably stronger — that's the main reason it exists.
Can I edit the notes afterwards?
Export to TXT or Markdown and edit anywhere — the Markdown drops cleanly into Notion, Obsidian or Google Docs. Timestamps survive the export so you can still jump back to the audio.
Does it work on my phone?
It's designed for desktop browsers — long files need memory and compute that phones don't like giving to a browser tab. The workflow that works: record on your phone, then drop the file on your laptop.
What do I get besides the transcript?
Auto-chapters with summaries and TXT/SRT/Markdown export for free. Study Packs (structured notes, glossary, quiz, Anki export) are part of Pro.