AitJazAudio and video to text, right on your PC.
AitJaz turns a meeting recording, interview, lecture or voice message into a Word document with paragraphs, timestamps and speaker labels. Recordings can be any length, and by default the file never leaves your computer.
7-day free trial · No account, no card · $20 one-time, no per-minute fees
Microsoft Store blocked or missing? Get the installer from this site (222 MB)
By downloading, you agree to the Terms of Service and Privacy Policy. How to install
Transcript · AitJaz
Speaker 1 · 00:00:04
Let's start. Deadlines first, then questions.
Speaker 2 · 00:00:11
The mockups still aren't approved. I'd move the launch to next week.
Speaker 1 · 00:00:19
Fine, let's move it. Send me your feedback on the mockups by Friday.
How it works.
Click the AitJaz icon in the tray, open Settings and pick the first section, “Audio & video to text”.
Pick a file
Leave “Sound source” on “File on this computer” and pick any audio or video, of any length. Choose the format to save: Word, text or subtitles.
Tick what you need
Timestamps, paragraphs by pauses and “Separate speakers”. Leave the number of people on Auto or set it from 2 to 6.
Start and wait
Progress shows in percent. When it's done, click “Speaker names…” and replace “Speaker 1” with a real name.
No recording? AitJaz records the meeting for you.
No need to take notes during the call. AitJaz records your microphone and everything your computer plays, which includes the other participants in Zoom, Teams or Google Meet. No bot joins the call, and by default the recording and text stay on your computer.
Choose “Meeting”
In “Sound source”, choose “Meeting (Zoom, Teams, Google Meet)”. The microphone and “Separate speakers” turn on by themselves; you can untick either one.
Start recording
During the call, click “● Start recording”. Once you've chosen “Meeting”, you can also just press Ctrl+Shift+X in any app. Tell the participants you are recording.
Stop and get the text
Click “Stop” on the recording panel or press Ctrl+Shift+X again. Lines are labeled “Speaker 1”, “Speaker 2”, and the sound (.m4a) is saved next to the text, so you can replay any unclear part.
The transcription language follows your dictation language, and the recording panel shows it, so you can fix a wrong one before the call ends. With no folder chosen, the meeting text and sound go to Downloads. The shortcut can be changed to Ctrl+Shift+U, D, G or H, or turned off: “Meeting” → “Recording hotkey”. Headphones give a cleaner recording. One recording can run up to 4 hours, then it stops and is transcribed.
What you get.
AitJaz saves the result to a folder you choose, or next to the recording if you leave the folder empty (meeting and computer sound recordings go to Downloads).
- Audio: mp3, m4a, aac, wav, ogg, opus (WhatsApp and Telegram voice messages), flac, wma, amr
- Video: mp4, mov, mkv, webm, avi and more; AitJaz takes the audio track
- Word (.docx): title, date and length, paragraphs, optional timestamps, each speaker in their own color
- Plain text (.txt) and subtitles (.srt), up to two lines per subtitle
To separate speakers, AitJaz downloads a voice model once (about 27 MB, internet needed).
Subtitles · AitJaz
1
00:00:04,000 --> 00:00:07,600
Speaker 1: Let's start.
Deadlines first, then questions.
2
00:00:11,000 --> 00:00:15,200
Speaker 2: The mockups still
aren't approved.
3
00:00:15,200 --> 00:00:18,400
I'd move the launch to next week.
Example of the .srt layout; the text is made up.
Private, and part of the same license.
Out of the box, recordings are transcribed on your computer and nothing is uploaded. If you prefer, you can connect a cloud service with your own API key: it stays off until you turn it on, and you pay that service directly.
- Meeting recording, interviews, lectures and video subtitles
- One $20 purchase covers dictation and transcription, no per-minute fees
- 7-day free trial with every feature, 14-day refund
Questions.
Is there a limit on recording length?
No. You can transcribe a five-minute stand-up or a recording several hours long. Progress shows in percent with the time left, and you can stop at any point with “Cancel”. Dictation keeps working while it runs. This is about existing files: a meeting or computer sound recording can run up to 4 hours.
How long does transcription take?
On our laptop (Intel Core i5-4210U, 12 GB RAM), one minute of Russian speech takes about 13–15 seconds, and speaker separation adds about 15 seconds per 5 minutes of recording. We haven't measured other languages yet. Times on your PC may differ, so check during the 7-day free trial.
Which languages does it support?
The same 18 languages as dictation, including English, Spanish, German, French, Russian and Kazakh. The speech language follows your dictation language; you can change it in “Audio & video to text”. For languages other than Russian, the speech model downloads once.
Do I need internet?
Only to download the app and, once, the speech model for languages other than Russian and the voice model (about 27 MB). After that it works offline, unless you turn on a cloud service yourself.
Can it record a Zoom, Teams or Google Meet meeting?
Yes. In “Sound source”, choose “Meeting” and click “Start recording”, or just press Ctrl+Shift+X in any app. AitJaz records your microphone and the computer sound, which carries the other participants, and after you stop it makes a transcript, split by speaker when “Separate speakers” is ticked (choosing “Meeting” ticks it for you). Nothing joins the call, so tell the participants you are recording.
Is the meeting audio kept?
Yes, when your microphone is recorded too (as it is with “Meeting” and with Ctrl+Shift+X): the sound is saved next to the text as an .m4a file, so you can listen again to any unclear part. If you don't choose a folder, both files go to Downloads.
Can it make subtitles for a video?
Yes. Pick the video and choose “Subtitles (.srt)”. The file contains timed phrases, up to two lines each.
What can't it do?
AitJaz doesn't write summaries or meeting notes: you get the full transcript. Files are transcribed one at a time. Speakers are separated automatically, so where people talked over each other, reread the labels.
How do I turn off autostart or uninstall AitJaz?
After setup, AitJaz starts with Windows, so dictation is ready as soon as your PC is on. To turn autostart off: AitJaz icon in the tray → “Settings” → “General” → uncheck “Start with Windows” → “Save”. AitJaz uninstalls like any other app: Windows Settings → Apps → Installed apps (Windows 10: Apps & features) → AitJaz → Uninstall (in Windows 11 the button is under “⋯”). This removes autostart too, and the uninstaller asks whether to delete downloaded speech models as well. Your settings and license key stay in the %APPDATA%\Aitjaz folder in case you reinstall; you can delete that folder by hand. This applies to the version from this site. For the Microsoft Store version, turn off autostart in Windows Settings → Apps → Startup; it uninstalls in the same Apps section.
Try it on your own recording.
Free for 7 days with every feature, no account and no card. If you like it, pay $20 once.
Microsoft Store blocked or missing? Get the installer from this site (222 MB)
By downloading, you agree to the Terms of Service and Privacy Policy.
Windows 10/11 (64-bit) · 222 MB installer