AI Meeting Notes

Record a lecture or meeting directly in your browser, or upload existing audio files. Scholarly automatically transcribes the audio, generates a summary and action items, and gives you an AI chat grounded in the transcript. It works equally well for a two-hour lecture and a team meeting — students get study-ready notes, and professionals get a clean record of what was decided and what needs doing.

Live Recording

Start a recording from the Create menu, the home page, or the sidebar. Once capture begins, a small bar appears at the bottom of the screen and follows you across the whole app.

• The capture bar -- It shows elapsed time and a level meter that moves while sound is arriving, and keeps Pause and Finish one click away. Open takes you back to the recording page whenever you want it.
• Leaving the page no longer stops capture -- Open a source, take notes, or start a chat while the recording keeps running. The bar comes with you.
• Audio is captured in your browser. No external software needed.
• Chunked uploads -- Audio uploads in small pieces continuously, so nothing is lost if your connection drops or your browser crashes.
• Pause and resume -- Pause the recording at any time and pick it back up when ready.
• One at a time -- Only one capture can run at once. Finish the active one before starting another.
• Max length -- Recordings can be up to 3 hours long, and you are told as you approach the limit.
• Click Finish when you are done. Processing begins automatically.

Live Transcript

Your words appear in the Transcript tab while you are still speaking.

• Lines land within seconds, cut at natural pauses, so each one reads as a finished phrase rather than a fragment.
• A status line under the newest line says what is happening right now: Listening, Transcribing, paused, or that no sound is reaching your microphone so nothing is being transcribed. If live text stalls, it says so and confirms your audio is still being recorded.
• The transcript follows the newest line on its own. Scroll back to read something earlier and a Jump to live button brings you back.
• Search the transcript at any point, during capture or after.
• When you press Finish, the full transcript is already on screen while the summary and action items are written.
• Long recordings keep their whole transcript. Use Show earlier lines or keep scrolling to load the rest.

Upload Audio

If you already have a recording, upload it instead.

• Supported formats: MP3, M4A, WAV, WebM, OGG.
• Upload from the Upload modal, the home page, or the flashcard creator.
• Each upload creates a dedicated recording page with the same transcription and AI features as a live recording.

Transcription and Processing

After recording or uploading, Scholarly processes the audio automatically:

• Timestamped transcript -- Every segment has a time marker. Click any segment to seek the audio player to that exact moment.
• AI-generated summary -- A concise overview of the recording content.
• Action items -- Pulled from the recording, so decisions and follow-ups from a meeting (or assignments mentioned in a lecture) don't get lost.
• Auto-generated title -- A descriptive title based on the content.
• Recording language -- A recording made in another language keeps that language. Its summary, topics, and title come back in the language that was spoken, not in English.

Recording Page

The recording page is your hub for reviewing and studying. At the top sit the title, the recording's status, and the date it was made. Everything else lives in one card below.

Title and Status

• Rename in place -- Click the large title and type. The new name is used everywhere the recording appears.
• The status line says where the recording stands -- recording, not processed, processing, ready, or processing failed -- alongside its length and date.

Audio Player

Once a recording is processed, its player sits between the title and the card.

• Playback speed control from 0.5x to 2x.
• Seek, skip forward, and skip back controls.
• Topic markers displayed on the progress bar so you can jump to key moments.

Tabs

One card holds all four tabs, so switching tabs swaps that card's contents.

• Transcript -- The live transcript while you record, and the full timestamped transcript afterwards. Click any segment to seek.
• Summary -- AI-generated overview of the recording.
• Action items -- Pulled from the recording, each with a timestamp so you can jump back to the moment it came up.
• AI Creations -- Create flashcards, quizzes, notes, and more from the recording, and see everything already made from it.

Before a recording is processed these tabs are not blank: each one tells you what will fill it once you finish.

Other Actions

• Download -- Download the original audio file (available in the three-dot menu).

Warnings and Problems

Every recording warning and failure shows up in the same place, as one notice at the top of the recording page that stays until it no longer applies -- not a toast that disappears before you can read it. Each notice carries the fix or a retry.

• Blocked or silent microphone -- You are told within seconds that nothing is being heard, along with what to check: the mic permission in the address bar, the right input, and whether another app has taken the microphone.
• Interrupted capture -- Scholarly watches for a computer going to sleep, a suspended tab, or a microphone that has stopped delivering audio. It restarts the microphone where it can, and either way it tells you roughly how much time went uncaptured instead of quietly handing you partial notes. That disclosure stays with the finished recording.
• Unsupported browser -- If your browser can't record audio, you are told before you start, with the option to upload a file instead.
• Failed processing -- Your audio is kept, and Try again picks up where it stopped.
• Abandoned recordings expire -- A capture left behind by a closed tab or a sleeping device ends on its own rather than blocking your next recording. Any audio already saved is kept, so you can open that recording and finish it.

AI Chat

Every completed recording has a dedicated chat panel.

• Ask questions and the AI searches your transcript to find relevant context before answering.
• Conversation suggestions are generated based on the recording content, so you can start studying immediately.

Flashcards from Audio

You can create flashcards directly from recordings:

• Open the flashcard creator from a completed recording.
• The flashcard creator also accepts audio file uploads directly.
• Completed recordings appear in the Library tab of the flashcard creator, so you can generate cards from any past recording.

Sharing

Share any recording with a link. Recipients can:

• Listen to the audio.
• Read the full transcript.
• View the summary and action items.
• Leave comments and reactions.

Limits

Recording and transcription do not consume an AI creation credit. Uploading an audio file uses one file-upload slot; recording live does not use a file-upload slot. Creating a separate artifact from the finished transcript — such as flashcards or a quiz — uses one AI creation credit for that artifact. See Plans and Limits for the full breakdown.