from openai import OpenAI
import gradio as gr
from transformers import pipeline
import numpy as np

transcriber = pipeline("automatic-speech-recognition", model="openai/whisper-base.en")
qa_model = pipeline("question-answering", model="distilbert-base-cased-distilled-squad")

def predict(message, history, api_key):
    print('in predict')
    client = OpenAI(api_key=api_key)
    history_openai_format = []
    if 0:
        for human, assistant in history:
            history_openai_format.append({"role": "user", "content": human})
            history_openai_format.append({"role": "assistant", "content": assistant})
    history_openai_format.append({"role": "user", "content": message})

    response = client.chat.completions.create(
        model='gpt-4o',
        messages=history_openai_format,
        temperature=1.0,
        stream=True
    )

    partial_message = ""
    for chunk in response:
        if chunk.choices[0].delta.content:
            print(111, chunk.choices[0].delta.content)
            partial_message += chunk.choices[0].delta.content
            yield partial_message

def chat_with_api_key(api_key, message, history):
    print('in chat_with_api_key')
    accumulated_message = ""
    for partial_message in predict(message, history, api_key):
        accumulated_message = partial_message
        history.append((message, accumulated_message))
        yield accumulated_message, history

def transcribe(audio):
    if audio is None:
        return "No audio recorded."
    sr, y = audio
    y = y.astype(np.float32)
    y /= np.max(np.abs(y))
    
    return transcriber({"sampling_rate": sr, "raw": y})["text"]




def clear_all():
    return None, "", ""

with gr.Blocks() as demo:
    #output = gr.Textbox(label="Output", elem_id="output")
    chatbot = gr.Chatbot(label="Chat Output")
    with gr.Row():
        api_key = gr.Textbox(label="API Key", placeholder="Enter your API key", type="password")
        message = gr.Textbox(label="Message")
    
    gr.Markdown("# Audio Transcription and Question Answering")
    with gr.Row():    
        audio_input = gr.Audio(label="Audio Input", sources=["microphone"], type="numpy")
        with gr.Column():
            transcription_output = gr.Textbox(label="Transcription")
            clear_button = gr.Button("Clear")
    state = gr.State([])
    

    def update_output(api_key, audio_input, state):
        print('in update_output')
        message = transcribe(audio_input)
        responses = chat_with_api_key(api_key, message, state)
        accumulated_response = ""
        for response, updated_state in responses:
            accumulated_response = response
            yield accumulated_response, updated_state

    btn = gr.Button("Submit")
    btn.click(update_output, inputs=[api_key, message, state], outputs=[chatbot, state])

    audio_input.stop_recording(
        fn=update_output,
        inputs=[api_key, audio_input, state],
        outputs=[chatbot, state]
    )
    
    clear_button.click(
        fn=clear_all,
        inputs=[],
        outputs=[audio_input, transcription_output, chatbot]
    )
    gr.HTML("""
    <style>
    #output {
        overflow-y: auto;
        max-height: 800px;
    }""")
    demo.launch()