Add B-K/umt5-thai-g2p-v2-0.5k (#1140)

* Add B-K/umt5-thai-g2p-v2-0.5k

* Add umt5_thaig2p test

Author: wannaphong

pythainlp/transliterate/core.py

@@ -122,6 +122,9 @@ def transliterate(
         * *tltk_ipa* - tltk, output is International Phonetic Alphabet (IPA)
         * *thaig2p_v2* - Thai Grapheme-to-Phoneme,
           output is IPA. https://huggingface.co/pythainlp/thaig2p-v2.0
+        * *umt5_thaig2p* - Thai Grapheme-to-Phoneme,
+          output is IPA, powered by UMT5.\
+          https://huggingface.co/B-K/umt5-thai-g2p-v2-0.5k
 
     :Example:
     ::
@@ -174,6 +177,8 @@ def transliterate(
         from pythainlp.transliterate.iso_11940 import transliterate
     elif engine == "thaig2p_v2":
         from pythainlp.transliterate.thaig2p_v2 import transliterate
+    elif engine == "umt5_thaig2p":
+        from pythainlp.translate.umt5_thaig2p import transliterate
     else:  # use default engine: "thaig2p"
         from pythainlp.transliterate.thaig2p import transliterate
 

pythainlp/transliterate/umt5_thaig2p.py

@@ -0,0 +1,34 @@
+# -*- coding: utf-8 -*-
+# SPDX-FileCopyrightText: 2016-2025 PyThaiNLP Project
+# SPDX-FileType: SOURCE
+# SPDX-License-Identifier: Apache-2.0
+"""
+umt5-thai-g2p-v2-0.5k 
+
+huggingface: https://huggingface.co/B-K/umt5-thai-g2p-v2-0.5k
+"""
+
+# Use a pipeline as a high-level helper
+from transformers import pipeline
+
+
+class Umt5ThaiG2P:
+    """
+    Latin transliteration of Thai words, using International Phonetic Alphabet
+    """
+
+    def __init__(self, device: str = "cpu"):
+        self.pipe = pipeline("text2text-generation", model="B-K/umt5-thai-g2p-v2-0.5k", device=device)
+
+    def g2p(self, text: str) -> str:
+        return self.pipe(text)[0]["generated_text"]
+
+
+_THAI_G2P = None
+
+
+def transliterate(text: str, device="cpu") -> str:
+    global _THAI_G2P
+    if _THAI_G2P is None:
+        _THAI_G2P = Umt5ThaiG2P(device=device)
+    return _THAI_G2P.g2p(text)

tests/extra/testx_transliterate.py

@@ -140,6 +140,8 @@ def test_transliterate(self):
         self.assertIsNotNone(transliterate("แมว", engine="thaig2p"))
         self.assertIsNotNone(transliterate("คน", engine="thaig2p_v2"))
         self.assertIsNotNone(transliterate("แมว", engine="thaig2p_v2"))
+        self.assertIsNotNone(transliterate("คน", engine="umt5_thaig2p"))
+        self.assertIsNotNone(transliterate("แมว", engine="umt5_thaig2p"))
         self.assertIsNotNone(transliterate("คน", engine="tltk_g2p"))
         self.assertIsNotNone(transliterate("แมว", engine="tltk_g2p"))
         self.assertIsNotNone(transliterate("คน", engine="tltk_ipa"))