app.py

import gradio as gr
from huggingface_hub import InferenceClient
import soundfile as sf
from transformers import pipeline
import torch

# Initialize the client for the text generation model
client = InferenceClient("Futuresony/future_ai_12_10_2024.gguf")

# Initialize the TTS pipeline from Huggingface
synthesizer = pipeline("text-to-speech", "Futuresony/output")

def respond(
    message,
    history: list[tuple[str, str]],
    system_message,
    max_tokens,
    temperature,
    top_p,
):
    # Prepare the messages for the chatbot
    messages = [{"role": "system", "content": system_message}]
    
    # Add history of previous conversation
    for val in history:
        if val[0]:
            messages.append({"role": "user", "content": val[0]})
        if val[1]:
            messages.append({"role": "assistant", "content": val[1]})
    
    messages.append({"role": "user", "content": message})

    response = ""

    # Generate the response from the model
    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
        yield response

    # Convert the generated text to speech
    speech = synthesizer(response)
    
    # Save the generated speech to a file
    sf.write("generated_speech.wav", speech["audio"], samplerate=speech["sampling_rate"])
    
    # Return both the text and the audio for playback
    return response, "generated_speech.wav"


# Create the Gradio interface with a textbox for the user to input a message
demo = gr.Interface(
    fn=respond,
    inputs=[
        gr.Textbox(label="Type your message here", placeholder="Enter message...", lines=2),
        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)"),
    ],
    outputs=[gr.Textbox(), gr.Audio()],
)

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