Everything you need to stop missing details

NoteMate transcribes your recordings, identifies each speaker, and turns the transcript into whichever summary format you need — automatically, in minutes.

Start free — no credit card See pricing

Core features

AI Transcription

Powered by AssemblyAI's Universal speech models. Upload a file or record directly — get a word-for-word transcript in minutes, not hours.

Speaker Identification

Automatically detects and labels each speaker. Rename them once and the name applies across your entire transcript instantly.

AI Summaries

GPT-powered summaries that capture decisions, action items, and key points — without you having to re-read the full transcript.

Flashcards

Turn a lecture into a deck you can review, built from your own notes. Every card keeps the timestamp it came from, so anything you half-remember takes one tap to hear again.

Quizzes that explain themselves

Multiple-choice questions from the same lecture, where every option carries an explanation — including the wrong ones, which say what makes them wrong rather than restating the answer.

Multiple Summary Formats

Lecture outline, seminar discussion, meeting notes, action items, interview summary, or brainstorm capture — choose the format that fits.

Timestamped Transcript

Every segment is timestamped and clickable. Jump to any moment in the recording with a single click — no scrubbing.

Notes in Your Language

Transcription covers 99 languages and detects the spoken one automatically. Your notes, flashcards and quizzes can be written in a different one — record an English lecture, revise in Chinese — with 11 languages to choose from.

Live preview while you record

Record in the browser or the iOS app and watch the transcript appear as you speak — or upload existing audio and video files. No bots, no integrations required.

Deep dive: Speaker Identification

Speaker identification is one of NoteMate's most powerful features. It automatically detects each person's voice and labels their contributions throughout the transcript.

Learn how speaker identification works →

Who uses NoteMate

Students

Record lectures, transcribe seminars, and get study notes without missing what's being said.

Professionals

Team meetings, client calls, one-on-ones — capture everything without staring at a laptop screen.

Researchers

Transcribe qualitative interviews with full speaker attribution and searchable, exportable transcripts.

Frequently asked questions

NoteMate supports MP3, MP4, M4A, WAV, WEBM, OGG, and most common audio and video formats. You can also record directly in the browser or using the iOS app — no file needed.

Most recordings are transcribed in a fraction of their actual length. A 60-minute lecture typically takes 2–5 minutes, depending on audio quality and current load.

NoteMate uses AssemblyAI's Universal speech models for transcription and speaker identification, and OpenAI's GPT models for AI summaries and title generation.

Yes. Transcription covers 99 languages and detects the spoken one automatically, and the output language is a separate setting — so a lecture recorded in English can produce notes in any of 11 languages: English, Chinese, Spanish, Hindi, Arabic, French, Portuguese, German, Japanese, Korean, and Vietnamese. Summaries, titles, flashcards and quizzes all follow it; the transcript stays in the language that was spoken, so you can always check a term against what was said.