metadata
license: mit
Hinglish Dump
Raw merged dump of Hinglish (hi-EN) datasets.
Subsets and features
Subsets:
- crowd_transliteration
- hindi_romanized_dump
- hindi_xlit
- hinge
- hinglish_norm
- news2018
_FEATURE_NAMES = [
"target_hinglish",
"source_hindi",
"parallel_english",
"annotations",
"raw_input",
"alternates",
]