disable %wor line by default

batchalign/cli/cli.py

@@ -141,6 +141,8 @@ def writer(doc, output):
               default=False, help="Use WhisperX instead of Rev.AI (default). Superceeds --whisper.")
 @click.option("--diarize/--nodiarize",
               default=False, help="Perform speaker diarization (this flag is ignored with Rev.AI)")
+@click.option("--wor/--nowor",
+              default=False, help="Should we write word level alignment line? Default to no.")
 @click.pass_context
 def transcribe(ctx, in_dir, out_dir, lang, num_speakers, **kwargs):
     """Create a transcript from audio files."""
@@ -151,7 +153,8 @@ def writer(doc, output):
         CHATFile(doc=doc, special_mor_=True).write(output
                                                    .replace(".wav", ".cha")
                                                    .replace(".mp4", ".cha")
-                                                   .replace(".mp3", ".cha"))
+                                                   .replace(".mp3", ".cha"),
+                                                   write_wor=kwargs.get("wor", False))
 
     asr = "rev"
     if kwargs["whisper"]:

batchalign/formats/chat/file.py

@@ -93,7 +93,7 @@ def __init__(self, path=None, doc=None, lines=None, special_mor_=False):
             self.__doc = doc
 
 
-    def write(self, path):
+    def write(self, path, write_wor=True):
         """Write the CHATFile to file.
 
         Parameters
@@ -102,13 +102,13 @@ def write(self, path):
             Path of where the CHAT file should get str.
         """
 
-        str_doc = self.__generate(self.__doc, self.__special_mor)
+        str_doc = self.__generate(self.__doc, self.__special_mor, write_wor=write_wor)
 
         with open(path, 'w') as df:
             df.write(str_doc)
 
     @staticmethod
-    def __generate(doc:Document, special=False):
+    def __generate(doc:Document, special=False, write_wor=True):
         utterances = doc.content
 
         def __get_birthdays(line):
@@ -125,7 +125,8 @@ def __get_birthdays(line):
                         extra += ":\t"+i.content
                     main.append(extra.strip())
             else:
-                main.append(generate_chat_utterance(i, special and doc.langs[0] == "eng"))
+                main.append(generate_chat_utterance(i, special and doc.langs[0] == "eng",
+                                                    write_wor=write_wor))
         main.append("@End\n")
 
         return "\n".join(main)

batchalign/formats/chat/generator.py

@@ -10,7 +10,7 @@
 # document[3].text = None
 # document[3].model_dump()
 
-def generate_chat_utterance(utterance: Utterance, special_mor=False):
+def generate_chat_utterance(utterance: Utterance, special_mor=False, write_wor=True):
     """Converts at Utterance to a CHAT string.
 
     Parameters
@@ -73,7 +73,7 @@ def generate_chat_utterance(utterance: Utterance, special_mor=False):
         result.append(f"%{'u' if special_mor else ''}gra:\t"+" ".join([f"{i.id}|{i.dep_id}|{i.dep_type}" for i in gras]))
 
     #### WOR LINE GENERATION ####
-    if has_wor:
+    if has_wor and write_wor:
         result.append("%wor:\t"+" ".join(wor_elems))
 
 

batchalign/version

@@ -1,3 +1,3 @@
-0.7.0-beta.1
-Mar 25rd, 2024
-Minor bug fixes
+0.7.0-beta.2
+Apr 05, 2024
+Disables %wor line by default in transcribe