app.py

import gradio as gr
from huggingface_hub import InferenceClient
import openai
from decouple import config
import win32com.client
import pythoncom

"""
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
"""

# Configure OpenAI for speech-to-text
client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")

def process_audio_and_respond(
    audio,
    history: list[tuple[str, str]],
    system_message,
    max_tokens,
    temperature,
    top_p,
):
    if audio is None:
        return "Please provide an audio input."
    
    # Convert speech to text using Whisper
    audio_file = open(audio, "rb")
    transcript = openai.Audio.transcribe("whisper-1", audio_file)
    user_message = transcript["text"]

    # Prepare messages for Zephyr
    messages = [{"role": "system", "content": system_message}]
    for user, assistant in history:
        if user:
            messages.append({"role": "user", "content": user})
        if assistant:
            messages.append({"role": "assistant", "content": assistant})
    messages.append({"role": "user", "content": user_message})

    # Get response from Zephyr
    response = ""
    for message in client.chat_completion(
        messages,
        max_tokens=max_tokens,
        stream=True,
        temperature=temperature,
        top_p=top_p,
    ):
        token = message.choices[0].delta.content
        response += token

    # Convert response to speech
    pythoncom.CoInitialize()
    speaker = win32com.client.Dispatch("SAPI.SpVoice")
    speaker.Speak(response)

    return user_message, response

"""
For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
"""
demo = gr.ChatInterface(
    process_audio_and_respond,
    additional_inputs=[
        gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
        gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
        gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
        gr.Slider(
            minimum=0.1,
            maximum=1.0,
            value=0.95,
            step=0.05,
            label="Top-p (nucleus sampling)",
        ),
    ],
)

if __name__ == "__main__":
    demo.launch()