diff --git a/README.md b/README.md
index 72dfdc5..46a8399 100644
--- a/README.md
+++ b/README.md
@@ -56,7 +56,7 @@ For reference, here's the time and memory usage that are required to transcribe
 
 ## Requirements
 
-* Python 3.9 or greater
+* Python 3.10 or greater
 
 Unlike openai-whisper, FFmpeg does **not** need to be installed on the system. The audio is decoded with the Python library [PyAV](https://github.com/PyAV-Org/PyAV) which bundles the FFmpeg libraries in its package.