Lecture transcription app

Two jobs hide inside one search

Job one is recorded-lecture transcription. You hit record during class, or your professor posted the lecture audio, and now you have a file. You want that file turned into searchable text you can read, skim, and quote. The input is audio you already have. The work is converting a recording. That is file transcription, and it's a real, legitimate category with tools built for it.

Job two is study notes by voice. You're reviewing the material — rereading the slides, working through the textbook, walking back from class with the lecture fresh — and you want to get your own thoughts down without typing. You talk; the words appear in your notes app. The input is your own voice, on purpose, in real time. That's dictation, and it's a different mechanism entirely. One processes a recording of someone else; the other types for you while you think.

Whisper by Remskill does the second job. It will not quietly record a room and transcribe it for you, and I'm not going to pretend it does — that's a recording tool's job, not a dictation tool's. What it does is remove the keyboard from the part where you write things up. For a lot of students, that second job is the one that actually eats the evening, which is the whole reason to care about it.

Press a hotkey, talk, your notes write themselves

The mechanic is boring, which is the point. You press a hotkey, you speak, you release, and the transcript pastes at your cursor — in Notion, in a Google Doc, in OneNote, in the Notes app, in whatever text field has focus. Whisper holds a short tail after you let go of the key so your last word doesn't get clipped. Because it pastes at the operating-system cursor, your notes app is just "any text box." It doesn't need a plugin and it doesn't care which app you're in.

For a student, that maps onto the actual work like this. You finish a reading, you summarise the chapter out loud in your own words, and the summary lands in your notes — far faster than typing it. You turn a slide into a question you'll quiz yourself on later. You dictate the gist of a lecture into your review doc while it's still fresh. A small capsule shows up while you speak so you know it's listening:

The hotkey is the one thing worth getting right up front. On Windows it's Ctrl+Space; on Mac it's Command+Option, a modifier-only push-to-talk you hold while speaking. Both are changeable in Settings if they clash with something you already use. (My younger daughter once told me a hotkey "didn't work" in her drawing app. It was a conflict, not a bug, which is how I learned the average person has no idea what a hotkey conflict even is. So now every hotkey is customisable.) If you've used dictation software for students before, this is the same muscle memory pointed at your notes.

Set it up in two minutes (Windows or Mac)

You need a Mac on Apple Silicon or a Windows 10-or-newer PC, a working microphone, and your notes app open. The whole local pipeline is free for any signed-in account, with no payment method asked for at sign-up. Here's the sequence.

The slow part is the model download, not the setup. Everything else is the four steps above. Once it's running, writing up a lecture stops being a typing task and becomes a talking task — which is a much better use of the half hour after class.

A study-notes workflow built around talking

The version of this that actually works for students isn't "dictate the whole lecture verbatim." It's reviewing out loud. After a lecture or a reading, you explain the idea back to yourself in plain words, and the explanation lands in your notes. The act of saying it is half the studying; getting it written down is the other half, and you've done both at once without touching the keyboard. The boring truth is that the best study tool is often just talking through the material — voice typing makes that produce notes instead of evaporating.

A few concrete moves. Summarise each section of a chapter in two or three spoken sentences, so your notes are your own paraphrase rather than copied text. Dictate self-quiz questions as you go, then turn them into flashcards later. Read a tricky paragraph, then say the confusion out loud — "I don't get why the second step depends on the first" — so your review doc holds your actual questions, not just the facts. A 90-minute lecture can become a 600-word summary in a few minutes of talking, which is a very different evening from typing it line by line.

If your subject leans on specific terms — anatomy, statutes, organic-chemistry names — local Whisper lets you add custom vocabulary and hotwords so the engine biases toward the words it would otherwise mangle. That doesn't make it a clinical or legal tool, and it won't certify anything; it just stops "myocardium" coming out as "my cardio." Get the words down fast by voice, then shape the structure — headings, bullets, links — with the keys you already use, because a dictation tool produces words, not your notes app's layout.

Local or cloud: which mode for study notes

For study notes, try local mode first. Your half-formed understanding of a topic, your guesses about what's on the exam, the lecture you only half-followed — none of that needs to leave your laptop. If your Mac is Apple Silicon or your PC is from the last few years, local handles everyday dictation without complaint, and cloud becomes the escape hatch rather than the default. It's also free, which matters more when you're a student than when you're not.

Here's how the three paths differ, because the app makes you pick and I'd rather you pick well:

• Local Parakeet — NVIDIA's TDT engine, around 600 MB, and the fastest local option — 5 to 10 times faster than Whisper on CPU. Covers English plus 24 other European languages, 25 in total. No translate-to-English. If you study in English or another European language, this is the quick, fully offline pick.
• Local Whisper — slower than Parakeet on the same machine, but the multilingual builds cover 99 languages and can translate to English, and it supports the custom vocabulary and hotwords that help with subject jargon. The English-only builds are English-only, not 99. Pick this for Chinese, Japanese, Korean, translation work, or term-heavy fields. Default English model is around 480 MB.
• Cloud (OpenAI, BYOK) — best accuracy and web access, using your own OpenAI key billed straight by OpenAI. Transcription runs on gpt-4o-mini-transcribe by default. Needs internet, so it's the one path that leaves your machine. The Cloud surface is part of Whisper Pro.

The boring truth is that for the kind of text most students dictate — summaries, questions, your own paraphrasing — local is plenty. Both local engines run fully on your machine with nothing sent to a server. Cloud earns its place when you want top-tier accuracy on a hard recording or you need the model to pull a fact off the web mid-sentence. For everyday note-writing, start local and only reach for cloud when local leaves you wanting.

Cleanup and structure: turning a spoken summary into real notes

Raw dictation comes out as a run-on. You say "okay so the mitochondria is the part that makes energy ATP and it has its own DNA which is weird, exam probably asks about that," and that's the unpunctuated wall any speech engine hands you. Cleaning it up is where the paths diverge.

Windows Voice Typing adds punctuation as you speak, and macOS Dictation handles basic punctuation when you say "comma" or "period." For heavier cleanup — stripping the "ums," fixing the run-ons, turning a spoken ramble into something you'd actually revise from — Whisper can run an AI pass. Say the activation phrase "Hey whisper" and the text gets enhanced before it lands. On a local model that runs through Ollama; in cloud mode it's gpt-5-mini by default.

For your notes app's own structure — headings, nested bullets, tags, links — the honest answer is that voice gets you the text and the app gets you the structure. Dictate the summary, then add the heading, the bullet, or the link the way you always do. No dictation tool conjures an outline into existence on command; anyone promising that is selling you a demo, not a Tuesday. Get the words down fast by voice, shape the notes with the keys you already know.

That same speak-then-clean flow pays off well beyond studying — you can also turn voice into clean text for any kind of note-taking with the one hotkey, so a paragraph you'd have typed becomes a few spoken sentences instead.

When you actually need lecture-audio transcription

Here's the line I promised to draw clearly. If the job is "I have a recording of the lecture and I want the whole thing turned into text," a dictation tool is the wrong category, and I'd be doing you a disservice to suggest otherwise. Whisper by Remskill types what you say at your cursor; it does not ingest an audio file of a professor and produce a transcript. For that job you want a recording-and-transcription workflow, and there's a dedicated guide for it.

Reach for true lecture-audio transcription when you want the lecturer's exact words — quotes for a paper, a verbatim record of a guest speaker, a class you missed and only have the audio for. Those tools take an audio or video file and return text, often with timestamps and sometimes with speaker labels. That's a genuinely different machine from a dictation hotkey, and it's the right one when the input is a recording rather than your own live voice. Read how to transcribe a recorded lecture for the workflow that actually fits that job.

And for the small stuff, the right tool is sometimes the free one already on your machine. If you only drop a two-line reminder into your notes, Windows Voice Typing (Windows key + H) and macOS Dictation cover it for nothing — both type into the focused field, though Windows' version needs an internet connection. A dedicated, system-wide tool earns its place when the notes get long, the subject gets multilingual or term-heavy, or you want one hotkey that behaves the same in your notes, your email, and your essay draft. Below that bar, use what's free. I'm not going to tell you to install an app for a one-line reminder.

If most of your writing is essays and assignments rather than quick notes, the logic in typing faster with your voice carries straight over, because the hotkey doesn't care whether the cursor is in a notes app or a word processor.

A lecture transcription app is really two tools wearing one search term. If you have a recording, transcribe it with a transcription tool. If you have your own understanding and an empty notes page, talk it out and let the words land where your cursor is. I wrote most of this guide by dictating into a text box that wasn't my notes app, with a tool that doesn't care which box it is, then pasted the lot into my outline. The studying, sadly, still has to happen in your own head.