Open source transcription software for the Mac: four picks, by what you need back

By Ryleigh Newman · Published October 10, 2026

Four open source tools transcribe audio and video files on a Mac without sending them anywhere. Buzz gives you subtitles, speaker identification and YouTube links. YapToText gives you plain text from a file you drop on it, free from the App Store. VoiceInk runs a queue of files through the model and cleanup you choose. whisper.cpp does it from Terminal. All four can run the speech model on your own Mac.

I make YapToText, so I have a stake here, and I put Buzz first because for recordings it does more than mine does. Every outside fact is from each project's own pages, read between October 3 and October 9, 2026. I do not rank them by accuracy.

This app is completely free. You don't have to make an account. There are no subscriptions, no difficult downloads, and nothing is locked. Made for Apple silicon Macs. Source on GitHub.

This guide in other languages: العربية, Čeština, Dansk, Deutsch, Ελληνικά, Español, Suomi, Français, עברית, हिन्दी, Magyar, Bahasa Indonesia, Italiano, 日本語, 한국어, Norsk, Nederlands, Polski, Português, Română, Русский, Slovenčina, Svenska, ไทย, Türkçe, Українська, Tiếng Việt, 简体中文, 繁體中文

On this page: 8 sections

1. At a glance
2. 1. Buzz: the most complete for recordings
3. 2. YapToText: plain text, with nothing to set up
4. 3. VoiceInk: a queue, with your choice of model
5. 4. whisper.cpp: the engine itself
6. How to check that one is really local
7. The short version
8. Questions

At a glance

From each project's own pages, read October 3 to 9, 2026. Check current prices and requirements before you rely on one.

1. Buzz: the most complete for recordings

Buzz describes itself as an app that transcribes and translates audio offline on your personal computer. It is made by Chidi Williams and is MIT licensed. You import a file or a YouTube link, pick a task, a language and a model, and run it.

• Output. TXT, SRT or VTT, with a word-level timings option.
• Tools. Speaker identification, speech separation for noisy audio, a watch folder, a command line, and a viewer with search.
• Models. Whisper, Whisper.cpp, Faster Whisper and others. Its FAQ calls Whisper.cpp the best option on Macs.
• Platforms. Mac, Windows and Linux. Its README says Buzz now requires Apple silicon on the Mac.

If you need subtitles or to know who said what, start here. The Buzz page has more detail.

2. YapToText: plain text, with nothing to set up

This is mine. It is a dictation app first, and its Transcribe a file card takes any audio or video file macOS can decode. Whisper Large v3 Turbo ships inside the App Store download, so there is no model to fetch, and the file is transcribed on the Mac.

• Output. Plain text, to copy or insert. Pick Raw Transcription for exactly what was said, or a mode such as Note for headings and bullets.
• Limits. One file at a time. No timestamps, subtitles or speaker labels.
• Price. Free, with no paid tier and no minute cap.
• Code. GPL-3.0, on GitHub.

It suits a lecture, an interview or a voice memo where the words are what you need. The file transcription guide walks through it.

This app is completely free. You don't have to make an account. There are no subscriptions, no difficult downloads, and nothing is locked. Made for Apple silicon Macs. Source on GitHub.

3. VoiceInk: a queue, with your choice of model

VoiceInk is open source under the GPL v3. The ready-built app is a paid download after a 7-day trial, and its site says you can build it from source for free. Open Transcribe in its sidebar, drag in files, choose a Mode, and click Start to process the queue one after another.

• Formats. wav, mp3, m4a, aiff, mp4, mov, aac, flac, caf, amr, ogg, opus and 3gp, among others.
• Models. Whichever batch-capable model the Mode uses: Parakeet, Whisper and more.
• Cleanup. An optional enhancement step, with Original and Enhanced tabs on the result.
• History. Each file gets a History entry with timings. Your source file is not changed.

The VoiceInk model guide covers which model to pick, and the pricing page has the licenses.

4. whisper.cpp: the engine itself

whisper.cpp is a C and C++ version of OpenAI's Whisper under the MIT License, and it is the engine YapToText is built on. Its README calls Apple silicon a first-class citizen. You clone it, download a model with its script, build it with CMake, and run whisper-cli on a file. The README says whisper-cli runs only with 16-bit WAV files and gives an ffmpeg command to convert anything else.

It asks the most of you for a first transcript, and gives you the model with nothing in between. The local Whisper guide has the commands and each model's size.

How to check that one is really local

• Turn Wi-Fi off. Transcribe a short file. If it works, the speech model is on your Mac.
• Read the license file. Open source means the code is public under a license that lets you read, build and change it.
• Look for cloud options. Buzz and VoiceInk can also use hosted services if you connect one. Leave those off to stay local.

The short version

• Subtitles and speakers. Buzz.
• Plain text, no setup. YapToText.
• Many files, your choice of model. VoiceInk.
• Terminal. whisper.cpp.
• All four. Can transcribe on your own Mac.

If I have a project's facts wrong, or missed one that belongs here, open an issue on GitHub. Every one gets read.

This app is completely free. You don't have to make an account. There are no subscriptions, no difficult downloads, and nothing is locked. Made for Apple silicon Macs. Source on GitHub.

About the author. Ryleigh Newman is a developer whose hands make typing difficult and who dictates every day with YapToText, the app this site describes. Facts about other apps come from their own sites, with the date they were read. More about the project.

Related

• The best transcription apps for Mac in October 2026, ranked for audio and video files
• Open-source dictation apps for the Mac, from someone who makes one
• A Buzz alternative for the Mac that types where you are
• Transcribe an audio file on a Mac, free, without uploading it
• Local Whisper on a Mac: what it needs, what it costs, and how to run it
• Which VoiceInk model to choose, from VoiceInk's own guidance
• A free MacWhisper alternative built for dictation first
• I cannot type, so I built a free, open-source dictation app for the Mac

Sources

• https://github.com/chidiwilliams/buzz (checked 2026-10-07)
• https://chidiwilliams.github.io/buzz/docs (checked 2026-10-07)
• https://chidiwilliams.github.io/buzz/docs/faq (checked 2026-10-07)
• https://github.com/ggml-org/whisper.cpp (checked 2026-10-07)
• https://tryvoiceink.com/docs/transcribe-audio-files (checked 2026-10-09)
• https://tryvoiceink.com/docs/local-models (checked 2026-10-09)
• https://tryvoiceink.com/docs/license-management (checked 2026-10-09)
• https://github.com/ryleighnewman/YapToText (checked 2026-10-10)
• https://apps.apple.com/us/app/yaptotext/id6786382289 (checked 2026-10-10)

Questions