Create app.py

app.py

@@ -0,0 +1,78 @@
+# app.py
+import os
+import gradio as gr
+from groq import Groq
+from gtts import gTTS
+import tempfile
+import whisper
+
+
+
+# Initialize Groq client
+GROQ_API_KEY = "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"
+client = Groq(api_key = GROQ_API_KEY)
+
+
+# Load the Whisper model
+whisper_model = whisper.load_model("base")  # You can use "small", "medium", or "large" depending on your preference
+
+# Function to convert audio to text using OpenAI Whisper
+def audio_to_text(audio_file):
+    audio = whisper.load_audio(audio_file)
+    audio = whisper.pad_or_trim(audio)
+    mel = whisper.log_mel_spectrogram(audio).to(whisper_model.device)
+    options = whisper.DecodingOptions(fp16=False)
+    result = whisper.decode(whisper_model, mel, options)
+    return result.text
+
+# Function to interact with Groq API and generate a response
+def interact_with_groq(user_input):
+    try:
+        chat_completion = client.chat.completions.create(
+            messages=[{"role": "user", "content": user_input}],
+            model="llama3-8b-8192",  # Use the appropriate model
+            stream=False,
+        )
+        return chat_completion.choices[0].message.content
+    except Exception as e:
+        return f"Error interacting with Groq API: {e}"
+
+# Function to convert text to speech using gTTS
+def text_to_audio(response_text):
+    tts = gTTS(response_text)
+    output_path = tempfile.NamedTemporaryFile(delete=False, suffix=".mp3").name
+    tts.save(output_path)
+    return output_path
+
+# Main function for the chatbot
+def voice_to_voice(audio_file):
+    try:
+        # Step 1: Convert voice input to text
+        print("Transcribing audio...")
+        transcribed_text = audio_to_text(audio_file)
+        print(f"Transcribed Text: {transcribed_text}")
+
+        # Step 2: Interact with LLM via Groq API
+        print("Getting LLM response...")
+        response_text = interact_with_groq(transcribed_text)
+        print(f"LLM Response: {response_text}")
+
+        # Step 3: Convert LLM response to audio
+        print("Generating audio response...")
+        audio_response = text_to_audio(response_text)
+        return transcribed_text, audio_response
+    except Exception as e:
+        return f"Error processing request: {e}", None
+
+# Gradio Interface
+interface = gr.Interface(
+    fn=voice_to_voice,
+    inputs=gr.Audio(type="filepath"),
+    outputs=[gr.Textbox(label="Transcribed Text"), gr.Audio(label="Response Audio")],
+    title="Real-Time Voice-to-Voice Chatbot",
+    description="A real-time voice-to-voice chatbot using Whisper for transcription, Groq API for LLM, and gTTS for audio response.",
+)
+
+# Launch the interface
+if __name__ == "__main__":
+    interface.launch()