Scripts
Use the output_script parameter to control the writing system for transcription output.
Supported Writing Scripts
| Script Name | Script Code | Supported Languages |
|---|---|---|
| Auto-detect | auto | All 207 languages |
| Latin/Roman | latin | All 207 languages |
| Devanagari | devanagari | 32 Indic languages: Awadhi, Bagheli, Bengali (Bangladesh), Bengali (India), Bhojpuri, Bodo, Chhattisgarhi, Dogri, Garo, Gondi, Haryanvi, Hindi, Ho, Kannada, Kashmiri, Khasi, Khond, Konkani, Magahi, Maithili, Malvi, Marathi, Marwari, Mizo, Mundari, Nepali, Rajasthani, Santali, Sindhi, Tamil, Tulu, Urdu |
| Bengali | bengali | All 32 Indic languages |
| Gurmukhi | gurmukhi | All 32 Indic languages |
| Gujarati | gujarati | All 32 Indic languages |
| Odia | oriya | All 32 Indic languages |
| Tamil | tamil | All 32 Indic languages |
| Telugu | telugu | All 32 Indic languages |
| Kannada | kannada | All 32 Indic languages |
| Malayalam | malayalam | All 32 Indic languages |
| Sinhala | sinhala | All 32 Indic languages |
| Thai | thai | All 32 Indic languages + Lao, Thai |
| Lao | lao | All 32 Indic languages + Lao, Thai |
| Burmese | burmese | All 32 Indic languages + Burmese |
| Khmer | khmer | All 32 Indic languages + Khmer |
| Tibetan | tibetan | All 32 Indic languages + Tibetan |
| Arabic | arabic | Arabic (Chadian), Arabic (Egyptian), Arabic (Global), Arabic (Gulf), Arabic (Iraqi), Arabic (Levantine), Arabic (Maghrebi), Hebrew, Kurdish (Sorani), Pashto, Persian, Sindhi, Urdu |
| Hebrew | hebrew | Arabic (Egyptian), Arabic (Global), Arabic (Gulf), Arabic (Iraqi), Arabic (Levantine), Hebrew, Persian, Urdu |
| Cyrillic | cyrillic | Core: Belarusian, Bulgarian, Russian, Serbian, Ukrainian. Can also represent: Albanian, Basque, Bavarian, Catalan, Croatian, Czech, Danish, Dutch, English (all variants), Finnish, French (all variants), German, Hungarian, Italian, Polish, Portuguese (all variants), Romanian, Slovak, Slovenian, Spanish (all variants), Swedish, Welsh |
| Mongolian | mongolian | Mongolian |
| Greek | greek | Core: Greek. Can also represent: Albanian, Basque, Bavarian, Belarusian, Bulgarian, Catalan, Croatian, Czech, Danish, Dutch, English (all variants), Finnish, French (all variants), German, Hungarian, Italian, Polish, Portuguese (all variants), Romanian, Russian, Serbian, Slovak, Slovenian, Spanish (all variants), Swedish, Ukrainian, Welsh |
| Armenian | armenian | Armenian |
| Georgian | georgian | Georgian |
| Korean (Hangul) | korean | Korean |
| Chinese (Simplified) | chinese | Chinese (Cantonese), Chinese (Mandarin). Note: Very limited capability for non-Chinese languages |
| Japanese | japanese | Japanese. Note: Very limited capability for non-Japanese languages |
| Latin Extended | latin-ext | All 207 languages (same as Latin, with enhanced diacritical support) |
Script Usage Examples
# Auto-detect script (recommended)
data = {"output_script": "auto"}
# Force Latin / Roman script
data = {"output_script": "latin"}
# Devanagari script
data = {"output_script": "devanagari"}