Qwen Scribe - Private local transcription and dictation for Apple Silicon
Show HN: Qwen Scribe – local transcription and dictation for Apple Silicon
Qwen Scribe delivers private, offline transcription and system-wide dictation specifically optimized for Apple Silicon Macs. Powered by Qwen3-ASR running on the Metal GPU, this tool ensures your audio and text data never leave your device, eliminating the need for API keys or cloud accounts. Users can drag-and-drop media files for instant transcription or utilize a seamless system-wide dictation feature by holding the right Command key. With support for automatic language detection, vocabulary hints, and SRT export, Qwen Scribe offers a secure, high-performance alternative to cloud-based speech services, keeping your workflow entirely local and private.
There is no account, API key, or Qwen Scribe cloud service. Audio and transcript text stay on the Mac.
- smalltorch
Stuff like this is what doctors should use instead of blasting patient data to a AI scribe who knows where. Im pretty sure that is illegal but it still takes place and I am pretty sure one of the reasons for the push to solve FHE...but its obviously way more practical and private to just do the processing locally.
- dandaka
How does it compare to models from handy.computer and whisprflow?
- jiehong
Congrats on the launch!
1. I see it uses python, but I was wondering if it would use mlx-swift instead. Just curious if you considered it.
2. So far, qwen ASR seem usually less well supported than parakeet or whisper so far in most dictation applications (voxtral is similar in that regard). Why do you think that is?
Thanks and good luck with your roadmap!