generate whitenoise

app.py

@@ -1,7 +1,29 @@
 import server as xvaserver
 import gradio as gr
+import requests
 
 def predict(input):
+	model_type = 'xVAPitch'
+	line = 'Test'
+	pace = 1.0
+	save_path = 'test.wav'
+	language = 'en'
+	base_speaker_emb = []
+	use_sr = 0
+	use_cleanup = 0
+
+	data = {
+	    'modelType': model_type,
+	    'sequence': line,
+	    'pace': pace,
+	    'outfile': save_path,
+	    'vocoder': 'n/a',
+	    'base_lang': language,
+	    'base_emb': base_speaker_emb,
+	    'useSR': use_sr,
+	    'useCleanup': use_cleanup,
+	}
+	requests.post('http://localhost:8008/synthesize', json=data)
 	return ''
 
 input_textbox = gr.Textbox(

server.py

@@ -5,7 +5,8 @@ import multiprocessing
 
 torch_dml_device = None
 
-if __name__ == '__main__':
+# if __name__ == '__main__':
+if True:
     server = None
     multiprocessing.freeze_support()