Update app.py

app.py

@@ -172,11 +172,17 @@ def Transcribe_V2(num_speakers, speaker_names, audio="temp_audio.wav"):
         # print(segments)
         conversation=[]
         for (i, segment) in enumerate(segments):
-      print(f"{i}, {segment["speaker"]}, {segments[i - 1]["speaker"]}")
-            if i == 0 or segments[i - 1]["speaker"] != segment["speaker"]:
-                if i != 0:
-                    conversation.append([GetSpeaker(segment["speaker"]), segment["text"][1:]]) # segment["speaker"] + ' ' + str(time(segment["start"])) + '\n\n'
-            conversation[-1][1] += segment["text"][1:]
+            # print(f"{i}, {segment["speaker"]}, {segments[i - 1]["speaker"]}, {}")
+            if not len(conversation):
+                conversation.append([GetSpeaker(segment["speaker"]), segment["text"].lstrip())
+            elif conversation[-1][0] == GetSpeaker(segment["speaker"]):
+                conversation[-1][1] +=  segment["text"].lstrip()
+            else:
+                conversation.append([GetSpeaker(segment["speaker"]), segment["text"].lstrip())
+            # if i == 0 or segments[i - 1]["speaker"] != segment["speaker"]:
+            #     if i != 0:
+            #         conversation.append([GetSpeaker(segment["speaker"]), segment["text"][1:]]) # segment["speaker"] + ' ' + str(time(segment["start"])) + '\n\n'
+            # conversation[-1][1] += segment["text"][1:]
         # return output
         return conversation, ("".join([f"{speaker} --> {text}\n" for speaker, text in conversation]))