Sindhi transcription: what to expect
Sindhi is a low-resource language for the speech model. Expect a rough draft with many corrections, and lines that sometimes slip into Urdu spelling. Slow, clear speech from one speaker gives the best draft; calls, gatherings and music make it much weaker.
The draft is a starting point. Click a line to hear it again and fix it in place, which is faster than typing a long recording from scratch.
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.
Sindhi script
Expect the transcript in the Perso-Arabic Sindhi alphabet used in Pakistan, which has many extra letters for sounds Urdu lacks, such as ٻ, ڄ, ڏ, ڳ and ڻ. These are the letters the model most often gets wrong or replaces with plain Urdu forms, so check them on every line.
Sindhi written in Devanagari, used by some speakers in India, is not produced. Exports are UTF-8 right-to-left text; YouTube shows RTL subtitles correctly, though some desktop subtitle editors misplace punctuation.
Urdu, English and regional speech
Sindhi speakers in Karachi and other cities switch freely to Urdu and English. Keep Sindhi selected when Sindhi dominates; auto-detect can label Sindhi as Urdu, which gives Urdu text. Regional varieties and the speech of the Indian Sindhi community will need more corrections.
What people record in Sindhi
Uploads include interviews for journalism and research, poetry readings including verses of Shah Abdul Latif, Sufi music sessions, lectures, and family voice messages. Sindhi communities in India, the Gulf, the UK and North America use it to keep a written record of conversations and speeches. Sung poetry gives a weaker draft than speech, so compare verses with a printed 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 talk or a recorded mehfil, see the Sindhi video to text page.