refactor: Add time delay before removing recording audios.

app.py

@@ -30,6 +30,8 @@ from transformers import (
     VitsTokenizer,
 )
 
+import time
+
 import gradio as gr
 
 import logging
@@ -90,6 +92,11 @@ def translate(text, srcLang, tgtLang):
     )
     return response.text
 
+def remove_file(file) {
+    time.sleep(600) # delay for 10 minutes
+    os.remove(file)
+}
+
 @app.get("/")
 def read_root():
     return{
@@ -125,7 +132,7 @@ async def translate_audio(
             generate_kwargs=generate_kwargs
         )
         
-        print(result)
+        # print(result)
         
         result_dict = {
             "transcribed_text": result['text'], 
@@ -166,6 +173,7 @@ async def translate_text(text: str,
         "tgtLang": tgtLang
     }
     return result_dict
+    
 @app.post("/synthesize/")
 async def synthesize(text: str):
     model = VitsModel.from_pretrained("facebook/mms-tts-tgl")
@@ -190,5 +198,5 @@ async def synthesize(text: str):
     return FileResponse(
         temp_file, 
         media_type="audio/wav", 
-        background=BackgroundTask(os.remove, temp_file)
+        background=BackgroundTask(remove_file, temp_file)
         )