Update app.py

app.py

@@ -10,11 +10,23 @@ logging.basicConfig(level=logging.INFO)
 api = TTS("tts_models/deu/fairseq/vits")
 count = 0
 
+
 def audio_tts(txt, audio_file):
-    return {'cat': 0.3, 'dog': 0.7}
+    global count
+    count += 1
+    if count > 50:
+        time.sleep(5)
+        os.system("rm -R /tmp/*")
+        count = 0
+        
+    api.tts_with_vc_to_file(txt,
+    speaker_wav=audio_file,
+    file_path="ouptut.wav")
+    return "ouptut.wav"
+    
 
-demo = gr.Interface(fn=image_classifier, inputs=[gr.Textbox(label="Input text TTS", value="Привет! Я Макс."),
-                                                gr.Audio(source="upload", type="filepath", label="Input audio")],
+demo = gr.Interface(fn=audio_tts, inputs=[gr.Textbox(label="Input text TTS", value="Привет! Я Макс."),
+                                          gr.Audio(source="upload", type="filepath", label="Input audio")],
                     outputs=gr.Audio(source="upload", type="filepath", label="Output audio"))
 
 demo.queue(concurrency_count=1).launch(show_error=True)