Start free

Lecture transcription.Upload audio, study from text.

Drop a lecture recording from your phone, a lecture-capture export, or a course platform URL. Get a timestamped transcript with the lecturer separated from student questions — plus an AI study summary.

Drop a file, or pick one

MP3 · WAV · M4A · MP4 · MOV · MKV · OGG · OPUS · FLAC · WEBM — up to 100 MB anonymously

Paste a link, we’ll fetch the audio

YouTube · TikTok · Vimeo · Twitter · SoundCloud · Spotify · 50+ more

Record straight from your browser

Sign up takes 30 seconds — recording opens right after, in the dashboard.

No card required~90s per 60-min fileSRT · VTT · DOCX · TXTFiles auto-deleted in 24h

↓ Watch what comes out

Hall recording in. Study-ready text out.

A lecture recording is rarely clean — the lecturer is on a lapel mic, student questions come from 40 feet away, and half the vocabulary is field-specific. We label every turn and pass your syllabus terms to the recognizer.

Phone recording · CHEM 304REC Lecturer + 2 students · 1:12:44
auto-detected en-US44.1 kHz mono · 96 kbps
~90s
Transcript · streaming93% accuracy
S1

So the rate-determining step here is the formation of the carbocation intermediate — that's why tertiary substrates react faster in SN1.

S2

Professor, would a polar protic solvent stabilize that intermediate?

S1

Good question. Yes — water or methanol stabilize the carbocation through solvation. That's exam-relevant, by the way.

S3

Is that in chapter seven or chapter eight?

93% on lapel-mic hall audioDOCX · SRT · TXT · JSON

↓ This is the dashboard

This is what loads when the job finishes.

Same layout as the real dashboard — Summary, full Transcript, Speakers tab, Exports. Key points and action items extracted automatically. Auto-tags on every job.

Try it on your own file — it's free

Three real options · honest comparison

Otter for Students. Handwritten notes. Or us.

Otter sells a student tier with a live-record app. Handwritten notes are free but lossy. We work with whatever audio you already captured — phone memo, lecture-capture export, or a Panopto download.

Option 01

Otter for Students

Live-record app with classroom mode. EN-only, 30-importing cap on the free tier.

RequiresOtter account + app open
Speaker labelsAcoustic, EN-tuned
LanguagesEnglish only
File importCapped on free tier
Custom vocabPaid tier only
Cost$8.33–20/mo subscription
Best forUndergrads in English-language lectures who want to record live on a laptop and don't need course-specific vocabulary.
Option 02

Transcription.Solutions

Drop the file. Or paste a Panopto/Echo360 URL. Course vocabulary supported, no subscription required.

RequiresJust the audio file
Speaker labelsLecturer + students split
Languages99, auto-detected
File importUp to 5 GB per job
Custom vocabFree, paste your syllabus
Cost · per min$0.03 pay-as-you-go
Best forStudents with course-specific terminology, non-English lectures, or recordings already sitting on their phone or in an LMS.
Option 03

Handwritten / typed notes

Free, but you only capture what you can write while listening. No search, no replay.

RequiresPen, paper, attention
Coverage~30% of spoken words
SearchNone
AccessibilityNone for hearing-impaired
ReviewLinear reread only
CostFree
Best forDiscussion seminars where engagement matters more than completeness, or short lectures you don't need to revisit.

Pricing and feature flags accurate as of 2026. Otter free-tier import caps change frequently — verify on otter.ai before assuming.

Specific to lectures

Three things that bite students on generic transcription tools.

Lecture audio breaks generic AI in predictable ways. Flip these before you upload.

What goes wrong

  1. 1Field jargon comes out phonetically. 'Carbocation' becomes 'car bow cation', 'eigenvalue' becomes 'eye gun value'. Generic models weren't trained on your syllabus.
  2. 2Student questions get attributed to the lecturer. Generic diarization assumes one dominant voice and merges quiet, distant speakers into it.
  3. 3Slide references vanish. 'As you can see in equation 4.2' has no anchor in the text — you can't tell what slide the lecturer was on.

What to flip here

  1. 1Paste the syllabus glossary or chapter list into Custom Vocabulary on the job form. Names of authors, technical terms, drug names — all pass to the recognizer as hints.
  2. 2Set the speaker model to Lecturer + audience. We bias toward one primary speaker and group all student turns under a separate label you can rename later.
  3. 3Turn on AI summary with topic tags. We extract bolded concepts and timestamps so 'Lecture 7 · 34:12 · SN1 vs SN2' is searchable across the semester.

Recommended job settings for lectures

Drop a lecture file and these flip on by default. Override per-job from the form.

Diarization
Lecturer + audience mode
Custom vocabulary
Paste syllabus / glossary
Language
Auto-detect · 99 supported
Filler words
Removed (um, uh, you know)
AI summary
Key concepts + topic tags
Export
DOCX · SRT · timestamped TXT

Accuracy · real-world numbers

95% on lapel-mic capture. Drops with hall reverb and back-row questions.

The ceiling is set by how the lecturer was miked, not the room. A wired lapel feeding the lecture-capture system is the best case. A phone propped on the desk in row 12 is the worst. Numbers below come from real student uploads.

95%
Lecture-capture export (Panopto, Echo360)

Direct feed from the lecturer's lapel mic. Minimal reverb, no crowd noise. Student questions still suffer if the room mic is off.

92%
Phone on the desk, front row

Single lecturer, moderate distance. Field-specific terms are the main error source — paste the syllabus into custom vocabulary.

86%
Phone mid-hall, Q&A sections

Reverberant 200-seat room. Lecturer stays clean; student questions from the back may merge into one speaker.

80%
Heavy accent + reverb + chalkboard taps

Worst case in our data — non-native English speaker, large lecture hall, board taps masking phonemes. Still usable for review.

Common questions

8 things students ask about lecture transcription.

01Is it legal to record my professor's lecture?+
Depends on jurisdiction and school policy. Most US schools allow personal-use recording with the lecturer's consent or as a disability accommodation. Check your university's recording policy and ask the professor — we can't give legal advice, but we can transcribe whatever you legally captured.
02Can I import directly from Panopto, Echo360, or Canvas?+
Paste the download URL on the job form. If the file is behind a login, download the MP4/M4A to your laptop first, then drop it here. We don't store your LMS credentials.
03How does it handle equations and math notation?+
Spoken math becomes prose — 'x squared plus two x' stays that way, not LaTeX. We don't OCR slides. For STEM lectures, pair the transcript with the slide PDF and you've got a complete study set.
04What about lecturers with strong accents?+
Multilingual models handle most accents well — accuracy drops 3-6 points vs a native speaker. Heavy reverb compounds it. Custom vocabulary with technical terms helps more than anything else, because that's where accented speech actually breaks transcription.
05Will the AI summary work as a study guide?+
It's a starting point, not a finished guide. You get key concepts, topic tags, and timestamps for each. Use it to jump to the spots in the transcript you want to reread — don't memorize the summary alone.
06Can I get SRT captions for disability services?+
Yes — SRT and VTT are included on every job, free. If your accessibility office needs human-verified captions for legal compliance, our AI output is a strong first pass but isn't a certified accommodation by itself.
07Can I search across all my lectures for a semester?+
Each transcript is searchable on its own page. We don't have a cross-job semester search yet — most students drop DOCX exports into Notion, Obsidian, or a Google Drive folder and search there.
08Who can see my recordings?+
Only you. Files are encrypted at rest, processed, and auto-deleted after 30 days unless you delete sooner. We don't train models on your audio and we don't share files with the professor or school.

Drop a lecture recording. Get a study-ready transcript.

30 free minutes every month. No card. Speaker labels, course-specific vocabulary, SRT captions, AI summary — all included.

Start free