app.py

import gradio as gr
from main import index, run, ingest_files
from gtts import gTTS
import os, time

from transformers import pipeline

p = pipeline("automatic-speech-recognition", model="openai/whisper-base")

"""Use text to call chat method from main.py"""

models = ["GPT-3.5", "Flan UL2", "Flan T5", "GPT-4"]

name = os.environ.get("name", "Rohan")

def add_text(history, text, model):
    print("Question asked: " + text)
    response = run_model(text, model)
    history = history + [(text, response)]
    print(history)
    return history, ""


def run_model(text, model):
    start_time = time.time()
    print("start time:" + str(start_time))
    response = run(text, model)
    end_time = time.time()
    # If response contains string `SOURCES:`, then add a \n before `SOURCES`
    if "SOURCES:" in response:
        response = response.replace("SOURCES:", "\nSOURCES:")
        # response = response + "\n\n" + "Time taken: " + str(end_time - start_time)
    print(response)
    print("Time taken: " + str(end_time - start_time))
    return response



def get_output(history, audio, model):

    txt = p(audio)["text"]
    # history.append(( (audio, ) , txt))
    audio_path = 'response.wav'
    response = run_model(txt, model)
    # Remove all text from SOURCES: to the end of the string
    trimmed_response = response.split("SOURCES:")[0]
    myobj = gTTS(text=trimmed_response, lang='en', slow=False)
    myobj.save(audio_path)
    # split audio by / and keep the last element
    # audio = audio.split("/")[-1]
    # audio = audio + ".wav"
    history.append(( (audio, ) , (audio_path, )))
    print(history)
    return history

def set_model(history, model, first_time=False):
    print("Model selected: " + model)
    history = get_first_message(history)
    index(model, first_time)
    return history


def get_first_message(history):
    history = [(None,
                "Hi! I am " + name + "'s Personal Assistant. Want " + name + " to answer your questions? Just Roar it!")]
    return history

def clear_audio(audio):
    return None

def bot(history):
    return history

def upload_file(files, history, model):
    file_paths = [file.name for file in files]
    print("Ingesting files: " + str(file_paths))
    text = 'Uploaded a file'
    if ingest_files(file_paths, model):
        response = 'Files are ingested. Roar now!'
    else:
        response = 'Files are not ingested. Please try again.'

    history = history + [(text, response)]
    return history

theme = gr.Theme.from_hub("snehilsanyal/scikit-learn")

theme.block_background_fill = gr.themes.colors.neutral.c100
theme.block_border_width = '2px'
theme.block_border_radius = '10px'


with gr.Blocks(theme=theme, title='Roar!') as demo:
    # Add image of Roar Logo from local directory
    gr.HTML('<img src="file/assets/logo.png" style="width: 100px; height: 100px; margin: 0 auto;border:5px solid orange;border-radius: 50%; display: block">')
    # Title on top in middle of the page
    gr.HTML("<h1 style='text-align: center;'>Roar - A Personal Assistant</h1>")

    chatbot = gr.Chatbot(get_first_message([]), elem_id="chatbot").style(height=500)

    with gr.Row():
        # Create radio button to select model
        radio = gr.Radio(models, label="Choose a model", value="GPT-4", type="value")
    with gr.Row():
        with gr.Column(scale=0.6):
            txt = gr.Textbox(
                label="Let's hear the roar!",
                placeholder="Enter text and press enter, or upload a file", lines=1
            ).style(container=False)

        with gr.Column(scale=0.2):
            upload = gr.UploadButton(label="Roar on a file", type="file", file_count='multiple', file_types=['docx', 'txt', 'pdf', 'html']).style(container=False)

        with gr.Column(scale=0.2):
            audio = gr.Audio(source="microphone", type="filepath", label="Let me hear your roar!").style(container=False)

    with gr.Row():
        gr.Examples(examples=['Roar it! What are you an expert of?', ' Roar it! What are you currently doing?',
                              'Roar it! What is your opinion on Large Language Models?'], inputs=[txt], label="Examples")

    txt.submit(add_text, [chatbot, txt, radio], [chatbot, txt], postprocess=False).then(
        bot, chatbot, chatbot
    )

    radio.change(fn=set_model, inputs=[chatbot, radio], outputs=[chatbot]).then(bot, chatbot, chatbot)

    audio.change(fn=get_output, inputs=[chatbot, audio, radio], outputs=[chatbot, audio], show_progress=True).then(
        bot, chatbot, chatbot, clear_audio
    )

    upload.upload(upload_file, inputs=[upload, chatbot, radio], outputs=[chatbot]).then(bot, chatbot, chatbot)

    set_model(chatbot, radio.value, first_time=True)


if __name__ == "__main__":
    demo.queue()
    demo.queue(concurrency_count=5)
    demo.launch(debug=True, favicon_path="file/assets/logo.png")