diff --git a/README.md b/README.md
index f775da3..419d555 100644
--- a/README.md
+++ b/README.md
@@ -30,6 +30,10 @@ If you find a better voice match for `tts-1` or `tts-1-hd`, please let me know s
 
 ## Recent Changes
 
+Version 0.17.2, 2024-07-01
+
+* fix -min image (re: langdetect)
+
 Version 0.17.1, 2024-07-01
 
 * fix ROCm (add langdetect to requirements-rocm.txt)
diff --git a/speech.py b/speech.py
index 687a8fc..2fd0e3e 100755
--- a/speech.py
+++ b/speech.py
@@ -16,7 +16,6 @@
 from openedai import OpenAIStub, BadRequestError, ServiceUnavailableError
 from pydantic import BaseModel
 import uvicorn
-from langdetect import detect
 
 @contextlib.asynccontextmanager
 async def lifespan(app):
@@ -405,6 +404,7 @@ def auto_torch_device():
         from TTS.tts.models.xtts import Xtts
         from TTS.utils.manage import ModelManager
         from TTS.tts.layers.xtts.tokenizer import split_sentence
+        from langdetect import detect
 
     if args.preload:
         xtts = xtts_wrapper(args.preload, device=args.xtts_device, unload_timer=args.unload_timer)