Add laonlp.transliterate.transliterate

docs/api/transliterate.rst

@@ -7,4 +7,5 @@ Modules
 -------
 
 .. autofunction:: lao2thai_script
-.. autofunction:: thai2lao_script
+.. autofunction:: thai2lao_script
+.. autofunction:: transliterate

laonlp/transliterate/__init__.py

@@ -18,8 +18,10 @@
     "lao2thai_script",
     "thai2lao_script",
     "lao2thai_transliteration",
-    "thai2lao_transliteration"
+    "thai2lao_transliteration",
+	"transliterate"
 ]
+from anyascii import anyascii
 
 # Naive Lao script to Thai script transliteration.
 # Data from https://github.com/google/language-resources/blob/master/lo/Laoo-Thai.txt
@@ -127,3 +129,14 @@ def thai2lao_script(text: str) -> str:
         else:
             new_text += c
     return new_text
+
+def transliterate(lao_word: str, engine:str="anyascii")->str:
+	"""
+    Lao transliterate
+
+    :param str sent: Lao text
+	:param str engine: engine. Now, LaoNLP support anyascii only.
+    :return: returns a Lao transliteration.
+    :rtype: str
+    """
+	return anyascii(lao_word)

requirements.txt

@@ -1,3 +1,4 @@
 pythainlp>=3.0.0
 huggingface_hub
-gensim
+gensim
+anyascii>=0.3.2

tests/test_transliterate.py

@@ -1,7 +1,7 @@
 # -*- coding: utf-8 -*-
 
 import unittest
-from laonlp.transliterate import lao2thai_script, thai2lao_script
+from laonlp.transliterate import lao2thai_script, thai2lao_script, transliterate
 
 
 class TestTransliteratePackage(unittest.TestCase):
@@ -10,3 +10,6 @@ def test_lao2thai_script(self):
 
     def test_thai2lao_script(self):
         self.assertIsNotNone(thai2lao_script("พาสาลาว"))
+
+    def test_transliterate(self):
+        self.assertEquals(transliterate("ສະຫວັນນະເຂດ"),"sahvannaekhd")