import gradio as gr
from huggingface_hub import InferenceClient
from gtts import gTTS
import os

# Inicializando o cliente da Hugging Face com o modelo de linguagem
client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")

# Função para gerar a resposta e o áudio
def respond(
    message,
    history: list[tuple[str, str]],
    system_message,
    max_tokens,
    temperature,
    top_p,
):
    messages = [{"role": "system", "content": system_message}]

    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_text = ""

    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_text += token

    # Convertendo o texto em fala
    tts = gTTS(response_text, lang='pt')
    audio_file = "response.mp3"
    tts.save(audio_file)

    return response_text, audio_file

# Interface do Gradio usando 'gr.Interface' para múltiplos tipos de saída
demo = gr.Interface(
    fn=respond,
    inputs=[
        gr.Textbox(label="User Input", placeholder="Digite sua mensagem aqui..."),
        gr.State([]),  # Histórico da conversa
        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(label="Chatbot Response"),
        gr.Audio(label="Response in Audio")
    ],
    title="Chatbot com TTS",
    description="Digite uma mensagem e o chatbot responderá com texto e voz."
)

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