Uzbek audio to text: what to expect
Uzbek has relatively little training data in the speech model. Clear speech from one person gives a draft that saves typing, but expect many corrections, especially in suffixes, names and numbers. Noisy calls, weddings and group conversations will need much more work.
Review with the synced player: the current line is highlighted, clicking a line jumps there, and your edits save automatically.
Upload the recording as it is: MP3, M4A, WAV, OGG, OPUS, FLAC or AAC, a voice note saved from WhatsApp or Telegram, or a video file such as MP4 or MOV. You can also record from your microphone in the browser. Long files are split and processed in parallel, so a ten-minute recording usually takes a minute or two.
Latin or Cyrillic
Uzbek is officially written in a Latin alphabet, while Cyrillic is still common among older people and in some media. The transcript will be in one of the two, and there is no switch to convert between them, so check the first lines and plan any conversion separately.
In Latin text, look at oʻ and gʻ: the draft may use a plain apostrophe, a different quote mark, or drop it. The apostrophe that marks a glottal stop, as in maʼlumot, is often missing. Fix these consistently if the text goes into print or subtitles.
Russian, Tajik and regional speech
Uzbek speakers in Tashkent often mix in Russian, and speakers in Samarkand and Bukhara may also speak Tajik. Keep Uzbek selected when Uzbek dominates; if a meeting is mostly Russian, select Russian instead. Auto-detect can label short Uzbek clips as Turkish, Kazakh or Tajik.
Regional speech from the Fergana Valley, Khorezm or Karakalpakstan sounds different from the Tashkent standard and needs more corrections. Uzbek from northern Afghanistan, traditionally written in Arabic script, is not produced in that script.
What people record in Uzbek
Uploads include voice messages from family members working in Russia, Korea or Turkey, interviews for research and journalism, university lectures, podcasts, and religious talks. Students use it for interviews they need to quote in a thesis.
Exports and next steps
Export TXT or DOCX for notes and documents, PDF for a fixed copy, SRT or VTT for subtitles, and CSV with one row per line and its start time. Timestamps and speaker names can be switched on or off in each export. Speaker labels are available on paid plans or with a Minute Pack.
For video files, such as a YouTube interview or a Telegram video, see the Uzbek video to text page.