Frequently asked questions
Qugora is system-wide voice input software for macOS and Windows. Here are the questions we hear most about corrections, privacy, pricing and release status.
What is Qugora?
Qugora is system-wide voice input (dictation) software for macOS and Windows. Speak into any app’s text field and an AI model drops the filler words, adds punctuation and types the result at your cursor. Its focus is fixing faster than typing: when a number, English word or name comes out wrong, swipe across it on the trackpad and pick a candidate or re-say it in place.
Is Qugora related to Quora?
No. Qugora (spelled q-u-g-o-r-a) is voice input software and has nothing to do with the Q&A site Quora; they are separate, unrelated products. Qugora’s official site is qugora.com.
Which operating systems does Qugora support?
Qugora supports macOS and Windows. On macOS it needs a Mac with Apple silicon (M-series); on Windows it needs Windows 10 or 11 (64-bit). Built-in auto-update tells you when a new version is out.
When can I download Qugora?
Not yet: Qugora is in pre-release, and downloads, sign-in and Pro subscriptions aren’t open. At launch, the Download page will offer the macOS and Windows installers directly. Prices on the Pricing page are planned and may change at launch.
How do I fix a wrong English word, number or name?
Swipe across the wrong words on the trackpad, then pick a candidate or re-say just those words, without going back to the keyboard. The model builds candidates from context and your personal lexicon, so the first one is usually right. If none fit, you can also describe the word (for example “Lin, as in two trees”) and only that span is replaced.
How does swipe-to-correct work?
Swipe-to-correct is three steps: swipe the wrong words, pick a candidate, or re-say them. Drag left-to-right across the words on the trackpad (close enough is fine) and a candidate bar pops up; picking one replaces the text in place. No trackpad? A mouse drag also brings up the candidate bar, and the keyboard is always there as a fallback.
How is Qugora different from built-in macOS dictation?
The biggest differences are which words you can fix and whether fixes are remembered. macOS Dictation puts a blue underline under words it is unsure about, and you can click one to choose another spelling; Qugora lets you swipe any span, not only the flagged words, then pick a candidate, re-say just those words or describe the word, and every fix goes into your personal lexicon so it is more likely right next time. Qugora also uses an AI model to remove filler words and add punctuation, while macOS Dictation is free with the system and supports on-device recognition for Mandarin (China mainland).
Does Qugora upload my audio?
Not by default. Speech recognition runs on your device, and your audio and text stay on your computer by default. Telemetry is counts and timings only, never text or audio, and you can turn it off; with the Qugora Cloud LLM we record token counts and don’t store request content.
Is my data used to train models?
No. We don’t use your audio, text or memory items to train models, and we don’t sell personal information. For cloud LLM requests we record token counts only, not content.
Does Qugora need an internet connection?
Basic voice input works offline because speech recognition runs on device. AI cleanup and candidates need a model: use Qugora Cloud (online), bring your own API key, or connect a local model.
Can I use my own LLM API key or a local model?
Yes, on both Free and Pro. Add your own API key or connect a local model in settings; the text then goes straight from your device to the provider you chose, never through us, and doesn’t count against your Qugora Cloud quota.
What’s the difference between Qugora Free and Pro?
Free is free forever and includes local recognition, the full swipe-to-correct flow, the personal lexicon and the local memory library, works with your own model key, and comes with 50k Qugora Cloud tokens a month. Pro is $30/month or $144/year, raises the Qugora Cloud quota to 3M tokens a month and adds memory sync. New accounts get a 14-day Pro trial with no card required, then return to Free automatically.
Which apps does Qugora work in?
Qugora works in pretty much any text field: WeChat, Feishu/Lark, DingTalk, ChatGPT (web and desktop), email, docs and browser forms. It works as a system-level input layer and types at your cursor, so apps need no integration.
What is the personal lexicon?
The personal lexicon is the set of spellings Qugora learns from your corrections, stored on your computer. Colleagues’ names, project codenames, English jargon and number formats are saved after one fix, and recognition and candidates prefer them next time. You can view, edit or delete entries any time.
How well does it handle mixed Chinese/English, numbers and names?
That’s exactly where Qugora focuses, but first-pass recognition will still make mistakes. So we put fast fixing first: swipe, pick, done, without touching the keyboard. Every fix teaches your lexicon, so the same word is more likely right next time.
How is Qugora related to Voconly?
Qugora is built on Voconly, an MIT-licensed open-source project. We added swipe-to-correct, the personal lexicon and the memory library on top of its system-wide voice input. Thanks to Voconly’s author and contributors.
Didn’t find your answer? Email hello@qugora.com and we’ll get back to you.
Next time a number comes out wrong, just swipe it.
Watch the correction demo See pricing
Pre-release: installers aren’t available yet. Downloads will appear here at launch.