BatchalignHK 0.7.22.post5__tar.gz → 0.7.22.post7__tar.gz

{batchalignhk-0.7.22.post5 → batchalignhk-0.7.22.post7}/BatchalignHK.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: BatchalignHK
-Version: 0.7.22.post5
+Version: 0.7.22.post7
 Summary: Python Speech Language Sample Analysis
 Author: Brian MacWhinney, Houjun Liu
 Author-email: macw@cmu.edu, houjun@cmu.edu

{batchalignhk-0.7.22.post5 → batchalignhk-0.7.22.post7}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: BatchalignHK
-Version: 0.7.22.post5
+Version: 0.7.22.post7
 Summary: Python Speech Language Sample Analysis
 Author: Brian MacWhinney, Houjun Liu
 Author-email: macw@cmu.edu, houjun@cmu.edu

{batchalignhk-0.7.22.post5 → batchalignhk-0.7.22.post7}/batchalign/formats/chat/file.py

@@ -127,8 +127,10 @@ class CHATFile(BaseFormat):
             elif str(i).strip() in ENDING_PUNCT:
                 continue
             else:
-                main.append(generate_chat_utterance(i, special and doc.langs[0] == "eng",
-                                                    write_wor=write_wor))
+                main.append(generate_chat_utterance(i,
+                                                    special and doc.langs[0] == "eng",
+                                                    write_wor=write_wor,
+                                                    merge_letters="yue" in doc.langs))
         main.append("@End\n")
 
         raw = "\n".join(main)

{batchalignhk-0.7.22.post5 → batchalignhk-0.7.22.post7}/batchalign/formats/chat/generator.py

@@ -11,7 +11,7 @@ import warnings
 # document[3].text = None
 # document[3].model_dump()
 
-def generate_chat_utterance(utterance: Utterance, special_mor=False, write_wor=True):
+def generate_chat_utterance(utterance: Utterance, special_mor=False, write_wor=True, merge_letters=False):
     """Converts at Utterance to a CHAT string.
 
     Parameters
@@ -30,6 +30,9 @@ def generate_chat_utterance(utterance: Utterance, special_mor=False, write_wor=T
     main_line = str(utterance)
     # last minut ecorrections
     # main_line = re.sub(r"<([\w ]+) \[\/", r"<\1> [/", main_line)
+    if merge_letters:
+        main_line = re.sub(r"([a-z]) ", r"\1", main_line)
+        main_line = re.sub(r"([a-z])([^a-z])", r"\1 \2", main_line)
     main_line = re.sub(r"«", "“", main_line)
     main_line = re.sub(r"»", "”", main_line)
     main_line = re.sub(r"—", "-", main_line)

{batchalignhk-0.7.22.post5 → batchalignhk-0.7.22.post7}/batchalign/pipelines/asr/tencent.py

@@ -189,40 +189,17 @@ class TencentEngine(BatchalignEngine):
             roman_cache = ""
             roman_cache_start = i.StartMs
             roman_cache_end = i.StartMs
+
             for j in i.Words:
                 word = j.Word
                 if self.__lang == "yue":
                     word = cc.convert(word)
-
                     word = self.replace_cantonese_words(word)
-
-                if self.is_roman(word):
-                    if roman_cache == "":
-                        roman_cache_start = (j.OffsetStartMs + start)
-                    roman_cache = roman_cache + word
-                    roman_cache_end = (j.OffsetEndMs + start)
-                else:
-                    if roman_cache != "":
-                        turn.append({
-                            "type": "text",
-                            "ts": roman_cache_start / 1000,
-                            "end_ts": roman_cache_end / 1000,
-                            "value": roman_cache
-                        })
-                    roman_cache = ""
-                    turn.append({
-                        "type": "text",
-                        "ts": (j.OffsetStartMs + start) / 1000,
-                        "end_ts": (j.OffsetEndMs + start) / 1000,
-                        "value": word
-                    })
-
-            if roman_cache != "":
                 turn.append({
                     "type": "text",
-                    "ts": roman_cache_start / 1000,
-                    "end_ts": roman_cache_end / 1000,
-                    "value": roman_cache
+                    "ts": (j.OffsetStartMs + start) / 1000,
+                    "end_ts": (j.OffsetEndMs + start) / 1000,
+                    "value": word
                 })
 
             turns.append({
@@ -232,7 +209,6 @@ class TencentEngine(BatchalignEngine):
         L.debug(f"Tencent done.")
 
         # Extract the text from the small volume parts for translation
-
         doc = process_generation({"monologues": turns},
                                 self.__lang_code, 
                                 utterance_engine=self.__engine)

{batchalignhk-0.7.22.post5 → batchalignhk-0.7.22.post7}/batchalign/pipelines/morphosyntax/ud.py

@@ -79,6 +79,8 @@ def handler(word, lang=None):
         target = word.text
     if not target:
         target = word.text
+    target = target.replace("」", "")
+    target = target.replace("「", "")
 
     # unknown flag
     unknown = False
@@ -145,7 +147,7 @@ def handler(word, lang=None):
         pos,target = verbform(pos,target,word.text)
         target = target.replace(',', 'cm')
 
-    target = re.sub(r'@\w$', '', target)
+    target = re.sub(r'@\w$', '', target).strip()
     return f"{'' if not unknown else '0'}{pos}|{target}"
 
 # POS specific handler

batchalignhk-0.7.22.post7/batchalign/version

@@ -0,0 +1,3 @@
+0.7.22-post.7
+September 28rd, 2025
+Remove spaces between letters for cantonese

batchalignhk-0.7.22.post5/batchalign/version

@@ -1,3 +0,0 @@
-0.7.22-post.5
-September 3rd, 2025
-Japanese parsing fixes