Marathi audio to text: what to expect
The speech model has much less Marathi training data than it has for Hindi or English. Clear speech from one or two people gives a usable draft that needs a careful read. Names, numbers and fast conversational speech need the most attention, and a noisy phone call or a recording with music underneath will need more correction.
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.
Devanagari and Marathi spelling
The transcript is written in Devanagari as used for Marathi, including ळ. Small spelling details are where you will spend most of your review time:
- Anusvara and nasal sounds may be written differently from your preferred spelling.
- Short and long vowels (इ and ई, उ and ऊ) can be swapped.
- English words may appear in Devanagari or in Latin letters.
- Some words may come out with Hindi spelling, since the two languages share the script.
Regional speech, Hindi and English
Marathi sounds different in Pune, Mumbai, Vidarbha, Marathwada and the Konkan coast. The model handles standard, clearly spoken Marathi best, and strong regional speech or rural recordings need more correction.
Everyday Marathi in cities mixes in English words and sometimes Hindi. Keep Marathi selected when most of the speech is Marathi. If a meeting is mainly in Hindi or English with a few Marathi sentences, choose that language instead. Auto-detect can label a short Marathi clip as Hindi, so setting the language yourself is safer.
What people transcribe in Marathi
Common files are college lectures and coaching classes, interviews for journalism and research, kirtan and pravachan recordings, podcasts, local news clips and family recordings. Marathi speakers in the US, the UK and the Gulf use it for video calls and voice notes from relatives in Maharashtra.
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 recorded play, see the Marathi video to text page.