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",
]