app.py

import gradio as gr
from huggingface_hub import InferenceClient
import torch
from TTS.api import TTS
import soundfile as sf

# Load TTS Model
tts_model = TTS("tts_models/en/ljspeech/tacotron2-DDC").to("cuda" if torch.cuda.is_available() else "cpu")

# Hugging Face LLM client (DeepSeek R1 7B)
client = InferenceClient("deepseek-ai/deepseek-r1-7b")


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 = ""

    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, None  # Yielding text response first

    # Generate speech from text response
    output_audio_path = "response.wav"
    tts_model.tts_to_file(text=response, file_path=output_audio_path)
    
    yield response, output_audio_path  # Yielding audio response


# Using gr.Blocks() instead of ChatInterface
with gr.Blocks() as demo:
    gr.Markdown("## DeepSeek R1 7B Chatbot with TTS")

    chatbot = gr.Chatbot()
    msg = gr.Textbox(label="User Input")
    
    system_msg = gr.Textbox(value="You are a friendly Chatbot.", label="System Message")
    max_tokens = gr.Slider(1, 2048, value=512, step=1, label="Max Tokens")
    temperature = gr.Slider(0.1, 4.0, value=0.7, step=0.1, label="Temperature")
    top_p = gr.Slider(0.1, 1.0, value=0.95, step=0.05, label="Top-p (Nucleus Sampling)")
    
    tts_audio = gr.Audio(type="filepath", label="TTS Output")

    def chat_fn(message, history):
        return respond(message, history, system_msg.value, max_tokens.value, temperature.value, top_p.value)

    msg.submit(chat_fn, inputs=[msg, chatbot], outputs=[chatbot, tts_audio])

demo.launch()