app.py

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

# Set up your TTS model (as before)
synthesiser = pipeline("text-to-speech", "Futuresony/output")

# Set up your text generation client
client = InferenceClient("Futuresony/future_ai_12_10_2024.gguf")

def respond(
    message,
    history: list[tuple[str, str]],
    system_message,
    max_tokens,
    temperature,
    top_p,
):
    # Generate text response from your model
    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

    # Convert the generated text into speech (Text-to-Speech)
    # Get speaker embedding (optional, if you want to control the speaker)
    embeddings_dataset = load_dataset("Matthijs/cmu-arctic-xvectors", split="validation")
    speaker_embedding = torch.tensor(embeddings_dataset[7306]["xvector"]).unsqueeze(0)

    # Generate speech from the text response
    speech = synthesiser(response, forward_params={"speaker_embeddings": speaker_embedding})

    # Save the speech to a file (you can play it on the fly or return it in other formats like MP3)
    sf.write("generated_speech.wav", speech["audio"], samplerate=speech["sampling_rate"])

    return response, "generated_speech.wav"
    # You can return the text along with speech if needed


# Create the Gradio interface
demo = gr.ChatInterface(
    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()