Update custom model files, README, and requirements

alignment.py

@@ -150,8 +150,8 @@ class ForcedAligner:
 
     # Offset compensation for Wav2Vec2-BASE systematic bias (in seconds)
     # Calibrated on librispeech-alignments dataset
-    START_OFFSET = 0.0  # Subtract from start times (shift earlier)
-    END_OFFSET = -0.0  # Add to end times (shift later)
+    START_OFFSET = 0.06  # Subtract from start times (shift earlier)
+    END_OFFSET = -0.03  # Add to end times (shift later)
 
     @classmethod
     def align(