Update app.py

app.py

@@ -1,12 +1,12 @@
 import gradio as gr
-import autocausalfrompretrained 
+from transformers import AutoTokenizer, AutoModelForCausalLM, pipeline, WhisperProcessor, WhisperForConditionalGeneration
+from gtts import gTTS
+import os
 
 class InteractiveChat:
-
-    whisper_processor = WhisperProcessor.from_pretrained("openai/whisper-large") 
-    whisper_model = WhisperForConditionalGeneration.from_pretrained("openai/whisper-large")
-
     def __init__(self):
+        self.whisper_processor = WhisperProcessor.from_pretrained("openai/whisper-large") 
+        self.whisper_model = WhisperForConditionalGeneration.from_pretrained("openai/whisper-large")
         self.zephyr_tokenizer = AutoTokenizer.from_pretrained("HuggingFaceH4/zephyr-7b-beta")
         self.zephyr_model = AutoModelForCausalLM.from_pretrained("HuggingFaceH4/zephyr-7b-beta", device_map="auto")
 
@@ -26,23 +26,22 @@ class InteractiveChat:
         return response
 
     def speak(self, text):
-        speech_client = SpeechClient() 
-        speech_client.synthesize(text)
+        tts = gTTS(text=text, lang='en')
+        tts.save("output.mp3")
+        os.system("mpg321 output.mp3")
 
-    def generate_response(self, input):
-      
-        # get transcription from Whisper
-      
-        response = self.get_zephyr_response(transcription)
-        
-        self.speak(response)
-        
-        return response
+# Create an instance of the InteractiveChat class
+chat = InteractiveChat()
+
+# Define a function that wraps the generate_response method
+def generate_response_fn(input_data):
+    return chat.generate_response(input_data)
 
+# Use the function in gr.Interface
 interface = gr.Interface(
-  gr.Audio(type="microphone"), 
+  gr.Audio(type="filepath"),  # Accept audio files
   gr.Textbox(),
-  self.generate_response
+  generate_response_fn  # Pass the function here
 )
 
 interface.launch()