Merge pull request #440 from jim60105/main

chore(writer): Join words without spaces for ja, zh

whisperx/transcribe.py

@@ -118,6 +118,7 @@ def cli():
                 f"{model_name} is an English-only model but receipted '{args['language']}'; using English instead."
             )
         args["language"] = "en"
+    align_language = args["language"] if args["language"] is not None else "en" # default to loading english if not specified
 
     temperature = args.pop("temperature")
     if (increment := args.pop("temperature_increment_on_fallback")) is not None:
@@ -174,7 +175,6 @@ def cli():
     if not no_align:
         tmp_results = results
         results = []
-        align_language = args["language"] if args["language"] is not None else "en" # default to loading english if not specified
         align_model, align_metadata = load_align_model(align_language, device, model_name=align_model)
         for result, audio_path in tmp_results:
             # >> Align
@@ -213,6 +213,7 @@ def cli():
             results.append((result, input_audio_path))
     # >> Write
     for result, audio_path in results:
+        result["language"] = align_language
         writer(result, audio_path, writer_args)
 
 if __name__ == "__main__":

whisperx/utils.py

@@ -123,6 +123,7 @@ TO_LANGUAGE_CODE = {
     "castilian": "es",
 }
 
+LANGUAGES_WITHOUT_SPACES = ["ja", "zh"]
 
 system_encoding = sys.getdefaultencoding()
 
@@ -283,7 +284,10 @@ class SubtitlesWriter(ResultWriter):
                 sstart, ssend, speaker = _[0]
                 subtitle_start = self.format_timestamp(sstart)
                 subtitle_end = self.format_timestamp(ssend)
-                subtitle_text = " ".join([word["word"] for word in subtitle])
+                if result["language"] in LANGUAGES_WITHOUT_SPACES:
+                    subtitle_text = "".join([word["word"] for word in subtitle])
+                else:
+                    subtitle_text = " ".join([word["word"] for word in subtitle])
                 has_timing = any(["start" in word for word in subtitle])
 
                 # add [$SPEAKER_ID]: to each subtitle if speaker is available