diff --git a/speech.py b/speech.py
index 4695055..1a16a9e 100755
--- a/speech.py
+++ b/speech.py
@@ -819,4 +819,5 @@ if __name__ == "__main__":
 
     # Use multiple workers for true concurrency (each worker = separate process with own GIL)
     # This prevents thread pool exhaustion and allows concurrent model loading
-    uvicorn.run(app, host=args.host, port=args.port, workers=4, timeout_keep_alive=300)
+    # Must use import string format for workers to function
+    uvicorn.run("speech:app", host=args.host, port=args.port, workers=4, timeout_keep_alive=300)