ahsk is a voice dictation app for Mac, Windows and iPhone, built by Jithin Kumar Palepu, who founded it and writes this site's guides and comparisons. It exists for one kind of person: someone who writes in English (or another second language) all day and loses time translating in their head, rereading and second-guessing tone. You speak in the language you think in; ahsk types clean text in the language you need.
- Jithin Kumar Palepu on LinkedIn
- On X
- On Bluesky
- On Threads
- Contact us
Support, billing, press and bug reports.
What we build
Three apps on one account. On the Mac, hold fn to dictate, tap it for hands-free, hold fn + Shift to translate as you speak, and record meetings as two tracks that come back as notes. Windows has the same product on Ctrl + Win. On iPhone, ahsk is a keyboard. Pro costs €9.99 a month for all three; the free tier is 10,000 dictated words with no card.
How it works, plainly
Speech is transcribed in the cloud by Whisper large-v3, the larger model, because accented speech is where small on-device models slip. A second model turns the transcript into finished text: punctuation, paragraphs, no filler words, and the tone of the mode you picked. That means ahsk needs a connection and has no offline mode. If your audio must never leave your laptop, a local app is the better pick, and our comparisons say so.
How we write about it
Every release is in the public changelog, including what broke. Comparison posts state the date prices were checked and link each vendor. When ahsk is the wrong choice for a reader, the post says that too. If you find something on this site that is out of date or wrong, write to us and we will fix it.
Try it on your own voice
Free for 10,000 words, no card. Say one paragraph you would normally type.
Download for MacRelated guides
Frequently Asked Questions
Who makes ahsk?
ahsk is built by Jithin Kumar Palepu, its founder, who also writes the guides and comparisons on this site.
Which platforms does ahsk run on?
Mac, Windows and iPhone, on one account and one subscription. There is no Android app.
Is my audio stored?
No. Dictation audio is sent for transcription and not stored on our servers, and it is never used to train models. The privacy policy lists every service that processes it.
