Languages
Every language, marked honestly.
4 languages are live in THATCHED Convert today. 19 more of India's languages are planned. Nothing below is promised by a date.
Live in Convert
All languages
“Open OCR model” means an open-source text-recognition model already exists for the language in Tesseract's published model list (checked 15 October 2026). Where none exists, supporting the language needs new training data, so we cannot promise it yet.
| Language | Script | Status | Open OCR model |
|---|---|---|---|
| తెలుగు Telugu | Telugu | Live | Yes |
| हिन्दी Hindi | Devanagari | Live | Yes |
| தமிழ் Tamil | Tamil | Live | Yes |
| English English | Latin | Live | Yes |
| অসমীয়া Assamese | Bengali-Assamese | Planned | Yes An open OCR model exists (Tesseract tessdata). We have not yet tested or offered it in Convert. |
| বাংলা Bengali | Bengali-Assamese | Planned | Yes An open OCR model exists (Tesseract tessdata). We have not yet tested or offered it in Convert. |
| बड़ो Bodo | Devanagari | Planned | No No open OCR model for this language in Tesseract's tessdata list (checked 15 October 2026). Supporting it needs new training data, so we cannot promise it yet. |
| डोगरी Dogri | Devanagari | Planned | No No open OCR model for this language in Tesseract's tessdata list (checked 15 October 2026). Supporting it needs new training data, so we cannot promise it yet. |
| ગુજરાતી Gujarati | Gujarati | Planned | Yes An open OCR model exists (Tesseract tessdata). We have not yet tested or offered it in Convert. |
| ಕನ್ನಡ Kannada | Kannada | Planned | Yes An open OCR model exists (Tesseract tessdata). We have not yet tested or offered it in Convert. |
| کٲشُر Kashmiri | Perso-Arabic | Planned | No No open OCR model for this language in Tesseract's tessdata list (checked 15 October 2026). Supporting it needs new training data, so we cannot promise it yet. |
| कोंकणी Konkani | Devanagari | Planned | No No open OCR model for this language in Tesseract's tessdata list (checked 15 October 2026). Supporting it needs new training data, so we cannot promise it yet. |
| मैथिली Maithili | Devanagari | Planned | No No open OCR model for this language in Tesseract's tessdata list (checked 15 October 2026). Supporting it needs new training data, so we cannot promise it yet. |
| മലയാളം Malayalam | Malayalam | Planned | Yes An open OCR model exists (Tesseract tessdata). We have not yet tested or offered it in Convert. |
| ꯃꯤꯇꯩꯂꯣꯟ Manipuri | Meetei Mayek | Planned | No No open OCR model for this language in Tesseract's tessdata list (checked 15 October 2026). Supporting it needs new training data, so we cannot promise it yet. |
| मराठी Marathi | Devanagari | Planned | Yes An open OCR model exists (Tesseract tessdata). We have not yet tested or offered it in Convert. |
| नेपाली Nepali | Devanagari | Planned | Yes An open OCR model exists (Tesseract tessdata). We have not yet tested or offered it in Convert. |
| ଓଡ଼ିଆ Odia | Odia | Planned | Yes An open OCR model exists (Tesseract tessdata). We have not yet tested or offered it in Convert. |
| ਪੰਜਾਬੀ Punjabi | Gurmukhi | Planned | Yes An open OCR model exists (Tesseract tessdata). We have not yet tested or offered it in Convert. |
| संस्कृतम् Sanskrit | Devanagari | Planned | Yes An open OCR model exists (Tesseract tessdata). We have not yet tested or offered it in Convert. |
| ᱥᱟᱱᱛᱟᱲᱤ Santali | Ol Chiki | Planned | No No open OCR model for this language in Tesseract's tessdata list (checked 15 October 2026). Supporting it needs new training data, so we cannot promise it yet. |
| سنڌي Sindhi | Perso-Arabic | Planned | Yes An open OCR model exists (Tesseract tessdata). We have not yet tested or offered it in Convert. |
| اردو Urdu | Perso-Arabic | Planned | Yes An open OCR model exists (Tesseract tessdata). We have not yet tested or offered it in Convert. |