Pashto transcription: what to expect
Pashto is a low-resource language for the speech model. Expect a rough draft with many corrections, particularly in names, places and fast conversation. One speaker talking clearly near the microphone works best. A crackly call, a crowded room or music underneath will make the draft much weaker.
Use the draft as a starting point. Clicking a line replays that moment, so you can check a word by ear and fix it on the spot.
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.
Pashto letters and right-to-left text
The transcript is written in the Pashto alphabet, which adds letters such as ټ, ډ, ړ, ښ, ږ, ځ and څ to the Arabic script. Check these letters, and the different forms of final ی (ي, ې, ۍ, ئ), which the model may swap or replace with Persian or Urdu forms.
Exports are UTF-8 right-to-left text. Some desktop subtitle editors place punctuation on the wrong side of an RTL line; YouTube displays it correctly.
Dialects and code-switching
Southern Pashto, around Kandahar and Quetta, and northern Pashto, around Peshawar, differ in pronunciation, for example in the sounds written ښ and ږ. The model writes standard spelling and may mishear dialect words, so fix them to match the speaker.
Speakers in Afghanistan often mix in Dari, and speakers in Pakistan mix in Urdu and English. Keep Pashto selected when Pashto dominates. Auto-detect can label Pashto as Persian, Urdu or Arabic, which produces the wrong text altogether.
What people record in Pashto
Uploads include news and radio segments, interviews for journalism and research, poetry and landay recitations, lectures and religious talks, and family voice messages. Pashtun communities in the Gulf, the UK, Germany and the US use it for calls with relatives. For anything legal, have a fluent person check the full text.
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 news clip, see the Pashto video to text page.