Update app.py

app.py

@@ -1,11 +1,14 @@
 import gradio as gr
 from huggingface_hub import InferenceClient
+import torch
+from TTS.api import TTS
+import soundfile as sf
 
-"""
-For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
-"""
-client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
+# Load TTS Model (supports multiple models)
+tts_model = TTS("tts_models/en/ljspeech/tacotron2-DDC").to("cuda" if torch.cuda.is_available() else "cpu")
 
+# Hugging Face LLM client
+client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
 
 def respond(
     message,
@@ -35,14 +38,16 @@ def respond(
         top_p=top_p,
     ):
         token = message.choices[0].delta.content
-
         response += token
-        yield response
+        yield response, None  # Yielding text response first
 
+    # Generate speech from text response
+    output_audio_path = "response.wav"
+    tts_model.tts_to_file(text=response, file_path=output_audio_path)
+    
+    yield response, output_audio_path  # Yielding audio response
 
-"""
-For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
-"""
+# Gradio Chat Interface with Audio Output
 demo = gr.ChatInterface(
     respond,
     additional_inputs=[
@@ -57,8 +62,11 @@ demo = gr.ChatInterface(
             label="Top-p (nucleus sampling)",
         ),
     ],
+    outputs=[
+        gr.Textbox(label="Generated Response"),
+        gr.Audio(type="filepath", label="TTS Output"),
+    ],
 )
 
-
 if __name__ == "__main__":
     demo.launch()