From 720f3d21258f52dace498cebdcf406d007cc9513 Mon Sep 17 00:00:00 2001 From: Isin Demirsahin Date: Wed, 4 Dec 2024 17:38:10 -0800 Subject: [PATCH] No public description PiperOrigin-RevId: 702919949 --- .../brahmic/grapheme_inventory.py | 1 + .../natural_translit/language_params/mr.py | 2 ++ .../romanization/testdata/mr_iso_nat.textproto | 15 +++++++++++++++ 3 files changed, 18 insertions(+) diff --git a/nisaba/scripts/natural_translit/brahmic/grapheme_inventory.py b/nisaba/scripts/natural_translit/brahmic/grapheme_inventory.py index d790ff3d..e4f532f2 100644 --- a/nisaba/scripts/natural_translit/brahmic/grapheme_inventory.py +++ b/nisaba/scripts/natural_translit/brahmic/grapheme_inventory.py @@ -131,6 +131,7 @@ SYMBOL = [c.Char(*args) for args in [ ['ind', '.', ph.SIL], + ['sgn', '-', ph.SIL], ['zwj', '+', ph.SIL], ['zwn', '|', ph.SIL], ]] diff --git a/nisaba/scripts/natural_translit/language_params/mr.py b/nisaba/scripts/natural_translit/language_params/mr.py index 0bca63f2..fc5e2281 100644 --- a/nisaba/scripts/natural_translit/language_params/mr.py +++ b/nisaba/scripts/natural_translit/language_params/mr.py @@ -46,6 +46,8 @@ _TXN_OPS = fl.FstList( + g2p.AI_TO_EH_LONG, + g2p.AU_TO_OH_LONG, g2p.A_TO_EC, g2p.VOCALIC_U, g2p.H_ASP, diff --git a/nisaba/scripts/natural_translit/romanization/testdata/mr_iso_nat.textproto b/nisaba/scripts/natural_translit/romanization/testdata/mr_iso_nat.textproto index 72ed8cdb..04c6156b 100644 --- a/nisaba/scripts/natural_translit/romanization/testdata/mr_iso_nat.textproto +++ b/nisaba/scripts/natural_translit/romanization/testdata/mr_iso_nat.textproto @@ -65,3 +65,18 @@ rewrite { input: "māṛʰā" output: "madha" } +rewrite { + rule: "ISO_TO_NAT" + input: "aisā" + output: "aisa" +} +rewrite { + rule: "ISO_TO_NAT" + input: "kannauja" + output: "kannauj" +} +rewrite { + rule: "ISO_TO_NAT" + input: "a-ēka" + output: "aek" +}