Qwen Scribe for MAC
Free, open-source transcription and dictation for Apple Silicon Macs. Runs Qwen3-ASR on your Mac, with no account and no cloud. Beta.
Qwen Scribe Overview
Qwen Scribe is a menu bar app with a local web interface that transcribes audio and video files and adds system-wide dictation on Apple Silicon Macs. It runs the Qwen3-ASR speech model on the Mac's own GPU through MLX. There is no account, no API key and no Qwen Scribe cloud service, so audio and transcripts stay on the Mac.
Hold a push-to-talk key in any text field, speak, and release: the text is transcribed locally and pasted at the cursor. For files, drag audio or video into the page, which runs in your browser at a local address. It is free, open-source software under the Apache-2.0 license. Version 0.3.0-beta.1 is a beta and focuses on dictation. It is ad-hoc signed and not notarized, so macOS blocks the first launch until you choose Open Anyway under System Settings > Privacy & Security. This release needs macOS 14 or later on Apple Silicon, plus Python 3.12 or newer, and ffmpeg for anything other than WAV files. The download is small because the Python dependencies and the model are fetched on first use.
Key Features
Qwen Scribe keeps speech-to-text on your own Mac.
- On-device: audio and transcripts stay on the Mac.
- Two models: 1.7B for accuracy, 0.6B for speed.
- Dictation: push-to-talk in any text field, or press to start and stop.
- File transcription: drag and drop audio and video, up to 4 GB.
- Languages: automatic detection, or any of the fourteen supported.
- Timestamps: word timestamps and SRT export.
- History: search, reopen, export and delete saved transcripts.
- Dictionary: names and terms you often dictate.
What’s New in Version 0.3.0-beta.1
Qwen Scribe 0.3.0-beta.1 is the dictation release.
- A dictation dictionary for names and terms
- A press-to-start, press-to-stop dictation mode
- Dictations can be kept out of the history
- Spoken "new line" and "new paragraph" in English dictation
- Replacements that turn a spoken phrase into the text you want
- The Fn (Globe) key as a push-to-talk key
Qwen Scribe FAQ
Is Qwen Scribe free?
Yes. It is free and open source under the Apache-2.0 license, and it needs no account or API key.
Is Qwen Scribe finished?
No. Version 0.3.0-beta.1 is a beta. Core transcription, history and dictation work, and the developer asks for feedback and careful testing.
Why does macOS block Qwen Scribe?
This build is ad-hoc signed and not notarized. Open it once, then approve it under System Settings > Privacy & Security > Open Anyway.
What does Qwen Scribe need?
An Apple Silicon Mac with macOS 14 or later and Python 3.12 or newer. ffmpeg is needed for audio and video other than WAV.
Does Qwen Scribe work on Intel Macs?
No. It needs an Apple Silicon Mac.
Does my audio leave the Mac?
No. Audio and transcript text stay on the Mac. The internet is used to download the model on first use.
How much memory do the models need?
The 1.7B model needs roughly 3.4 GB of unified memory and the 0.6B model roughly 1.2 GB, on top of normal app use.
How does dictation work?
Hold the push-to-talk key in any text field, speak and release. The app needs Microphone, Input Monitoring and Accessibility permission for this.
Is there a file size limit?
Yes. Each file can be up to 4 GB, and one job briefly needs twice the file size in free disk space.
Where does the Qwen Scribe download on qwen-scribe.FULLMAC.app come from?
The download button leads to the official Qwen Scribe releases on GitHub at https://github.com/VladUZH/qwen-scribe/releases. qwen-scribe.FULLMAC.app does not change or repackage the app.
Similar Apps
LlamaportLlamaport is a small Mac app for running and downloading local GGUF language models. It wraps llama-server from the llama.cpp project: it lists the models in your models folder, launches one with the full command shown before you press start, shows what it costs while it runs, and fetches new models from Hugging Face with downloads that can resume after a kill or a reboot. It reads GGUF headers directly, so the list shows the real quantisation, context length and mixture-of-experts structure rather than guesses from the file name.AI
UnfoldlyUnfoldly is a local AI search app for Mac that finds files by what you remember about them, not by their names. You choose the folders and files it should index, and it builds a searchable memory of them on your Mac: text from documents, what is shown in images, moments in videos and the content of audio recordings. Then you describe what you are looking for, such as a photo of your dog or the PDF that mentioned a pricing plan, and ask follow-up questions about the files it finds: summarize a document, ask about a video or compare notes.AI
Claude Usage MonitorClaude Usage Monitor is a free, open-source menu bar app that shows how much of your Claude.ai usage you have left. Its menu bar icon turns from green to orange to red as you approach your limits, and the text next to it shows either your current session percentage or, while you are working, an estimate of the time left at your present pace together with your weekly usage. Click it to open a small dashboard with progress bars for the current session, your weekly limits across all models and, on Max plans, Sonnet-only usage, each with a countdown to its next reset.AI
M-CourtyardM-Courtyard is a desktop app for fine-tuning large language models locally on an Apple Silicon Mac, without writing code. You import your documents, turn them into a training dataset, run a LoRA fine-tune with Apple's MLX through mlx-lm, test the result in a built-in chat and export it. The project says no data leaves your Mac.AI
EnviousWisprEnviousWispr is a dictation app for the Mac menu bar. Press a keyboard shortcut in any app, speak, and your words are transcribed on your Mac and pasted where you were typing. You can choose between two speech models: Parakeet Ultra, the fast default for 25 European languages, and Whisper, which covers more than 99 languages and detects the language automatically. It works offline and needs no account or subscription.AI
Agent Teams AIAgent Teams AI is a desktop app for coordinating teams of AI agents, for coding and broader project work. You create a team with one prompt and give its members roles. The agents then work in parallel, message each other, manage their own tasks and review each other's work, while tasks move across a live kanban board. You can send a message to any agent, comment on a task, and approve or reject the code each task changed.AI
FULLMAC.app does not host this app. Download buttons lead to the developer’s own pages.