init

app.py

@@ -8,21 +8,29 @@ nltk.download('averaged_perceptron_tagger_eng')
 
 # Get device
 device = 'auto'
-model = TTS(language='EN', device=device)
+model = TTS(language='EN_NEWEST', device=device)
 speaker_ids = model.hps.data.spk2id
 
 
-def inference(text: str):
+def inference(text: str, speed: float):
     output_path = 'en-us.wav'
-    model.tts_to_file(text, speaker_ids['EN-US'], output_path, speed=1.0)
+    model.tts_to_file(text, speaker_ids['EN-US'], output_path, speed=speed)
     return output_path
 
 
 if __name__ == "__main__":
     demo = gr.Interface(
+        title='Text-to-Speech',
+        description='Convert English text to speech',
         fn=inference,
         inputs=[
-            gr.Textbox(),
+            gr.Textbox(label='Text to Synthesize'),
+            gr.Slider(minimum=0.5, maximum=3.0, value=1.0, label='Speed'),
+            gr.Dropdown(
+                label='Speaker',
+                choices=['EN', 'EN-US', 'EN-BR', 'EN_INDIA', 'EN-AU', 'EN-Default'],
+                default='EN-US'
+            )
         ],
         outputs=[
             gr.Audio()