Update sv.py

sv.py

@@ -185,16 +185,16 @@ def parse_time(time_str):
     return int(h) * 3600 + int(m) * 60 + float(s)
 
 
-def format_time(seconds, use_short_format=True):
+def format_time(seconds, use_short_format=True, always_use_seconds=False):
     if isinstance(seconds, datetime.timedelta):
         seconds = seconds.total_seconds()
 
     minutes, seconds = divmod(seconds, 60)
     hours, minutes = divmod(int(minutes), 60)
 
-    if use_short_format or (hours == 0 and minutes == 0):
-        return f"{seconds:05.3f}s"
-    elif hours == 0:
+    if always_use_seconds or (use_short_format and hours == 0 and minutes == 0):
+        return f"{seconds:06.3f}s"
+    elif use_short_format or hours == 0:
         return f"{minutes:02d}:{seconds:06.3f}"
     else:
         return f"{hours:02d}:{minutes:02d}:{seconds:06.3f}"
@@ -348,9 +348,9 @@ def process_audio(audio_path, language="yue", fs=16000):
     # Format the results
     formatted_text = ""
     for speaker, start, end, duration, text in results:
-        start_str = format_time_with_leading_zeros(start)
-        end_str = format_time_with_leading_zeros(end)
-        duration_str = format_time_with_leading_zeros(duration)
+        start_str = format_time(start, always_use_seconds=True)
+        end_str = format_time(end, always_use_seconds=True)
+        duration_str = format_time(duration, always_use_seconds=True)
         speaker_num = "1" if speaker == "SPEAKER_00" else "2"
         line = f"{start_str} - {end_str} ({duration_str}) Speaker {speaker_num}: {text}"
         formatted_text += line + "\n"