diff --git a/TTS/bin/synthesize.py b/TTS/bin/synthesize.py
index 99fc2a58..5ff1181f 100755
--- a/TTS/bin/synthesize.py
+++ b/TTS/bin/synthesize.py
@@ -8,10 +8,6 @@ from argparse import RawTextHelpFormatter
 # pylint: disable=redefined-outer-name, unused-argument
 from pathlib import Path
 
-from TTS.api import TTS
-from TTS.utils.manage import ModelManager
-from TTS.utils.synthesizer import Synthesizer
-
 description = """
 Synthesize speech on command line.
 
@@ -339,6 +335,11 @@ def main():
     if not any(check_args):
         parser.parse_args(["-h"])
 
+    # Late-import to make things load faster
+    from TTS.api import TTS
+    from TTS.utils.manage import ModelManager
+    from TTS.utils.synthesizer import Synthesizer
+
     # load model manager
     path = Path(__file__).parent / "../.models.json"
     manager = ModelManager(path, progress_bar=args.progress_bar)