Update app.py import os import whisper from gtts import gTTS from groq import Groq import gradio as gr # Initialize Whisper model model = whisper.load_model("base") # Initialize Groq API (set your GROQ_API_KEY in the environment) client = Groq(api_key=os.environ.get("GROQ_API_KEY")) # Step 1: Transcribe Audio (Speech-to-Text using Whisper) def transcribe_audio(audio_path): result = model.transcribe(audio_path) return result['text'] # Step 2: Interact with LLM (Groq API) def interact_with_llm(user_input): chat_completion = client.chat.completions.create( messages=[ { "role": "user", "content": user_input, } ], model="llama3-8b-8192", stream=False, ) response = chat_completion.choices[0].message.content return response # Step 3: Convert Text to Speech using gTTS def text_to_speech(text): tts = gTTS(text, lang="en") audio_file = "response.mp3" tts.save(audio_file) return audio_file # Combined workflow: Transcribe -> Interact with LLM -> Convert to Speech def chatbot(audio): # Step 1: Transcribe Audio to Text transcription = transcribe_audio(audio) # Step 2: Get LLM response based on transcription llm_response = interact_with_llm(transcription) # Step 3: Convert LLM response to audio (text-to-speech) audio_output = text_to_speech(llm_response) return transcription, llm_response, audio_output # Gradio Interface setup interface = gr.Interface( fn=chatbot, inputs=gr.Audio(type="filepath", label="Speak into the microphone"), outputs=[ "text", # Transcription output "text", # LLM response output gr.Audio(type="filepath", label="Response Audio") # Final audio output ], live=True, title="Real-Time Voice-to-Voice Chatbot", description="Talk to an AI in real-time! Speak into the microphone, get a response, and hear it back.", ) # Launch Gradio app interface.launch()
verified
