app.py

import os
import openai
import torch 

import gradio as gr
import pytube as pt
from transformers import pipeline
from huggingface_hub import model_info

openai.api_key = os.getenv('OPEN_AI_KEY')
hf_t_key = ('HF_TOKEN_KEY')

MODEL_NAME = "openai/whisper-small" 
lang = "en"

device = 0 if torch.cuda.is_available() else "cpu"
pipe = pipeline(
    task="automatic-speech-recognition",
    model=MODEL_NAME,
    chunk_length_s=30,
    device=device,
)

pipe.model.config.forced_decoder_ids = pipe.tokenizer.get_decoder_prompt_ids(language=lang, task="transcribe")

def transcribe(microphone, file_upload):
    warn_output = ""
    if (microphone is not None) and (file_upload is not None):
        warn_output = (
            "WARNING: You've uploaded a recorded audio file . "
            "The recorded file from the microphone uploaded, transcribed and immediately discarded.\n"
        )

    elif (microphone is None) and (file_upload is None):
        return "ERROR: You have to either use the microphone or upload an audio file"

    file = microphone if microphone is not None else file_upload

    text = pipe(file)["text"]

    return warn_output + text


def _return_yt_html_embed(yt_url):
    video_id = yt_url.split("?v=")[-1]
    HTML_str = (
        f'<center> <iframe width="500" height="320" src="https://www.youtube.com/embed/{video_id}"> </iframe>'
        " </center>"
    )
    return HTML_str


def yt_transcribe(yt_url):
    yt = pt.YouTube(yt_url)
    html_embed_str = _return_yt_html_embed(yt_url)
    stream = yt.streams.filter(only_audio=True)[0]
    stream.download(filename="audio.mp3")

    text = pipe("audio.mp3")["text"]

    return html_embed_str, text


def predict(message, history):
    history_openai_format = []
    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 = openai.ChatCompletion.create(
        model='ft:gpt-3.5-turbo-1106:2292030-peach-tech::8cxzbHH4',
        messages= history_openai_format,
        temperature=1.0,
        stream=True
    )

    partial_message = ""
    for chunk in response:
        if len(chunk['choices'][0]['delta']) != 0:
            partial_message = partial_message + chunk['choices'][0]['delta']['content']
            yield partial_message

A1 = gr.ChatInterface(predict,
                title="COLLEAGUE",
                description="The Consummate AI Productivity Companion Suite for Freelancers, Entrepreneurs, and Professionals, that Chats, Writes, Transcribes, and Creates, Built By Peach State Innovation and Technology. Select The Corresponding Tab For Tool Accessibility",
                textbox=gr.Textbox(placeholder="Enter your question/prompt here..."),
                theme= gr.themes.Glass(primary_hue="neutral", neutral_hue="slate"),
                retry_btn=None,
                clear_btn="Clear Conversation")


A3 = gr.load(
             "models/Salesforce/blip-image-captioning-large",
              title=" ",
              description="Take a Photo or Upload Any Type of Imagery, I'll Give You Its Description",
              outputs=[gr.Textbox(label="I see...")],
              theme= gr.themes.Glass(primary_hue="neutral", neutral_hue="slate"))

A4 = gr.load(
             "models/stabilityai/stable-diffusion-xl-base-1.0",
              inputs=[gr.Textbox(label="Enter Your Image Description")],
              outputs=[gr.Image(label="Image")],
              title=" ",
              description="Bring Your Imagination Into Existence, Create Unique Images With COLLEAGUE",
              allow_flagging="never", 
              examples=["A gigantic celtic leprechaun wandering the streets of downtown Atlanta","A child eating pizza in a Brazilian favela"])

A5 = gr.HTML(
            value=("""
                    <iframe
	                src="https://peachtechai-colleague-scribe.hf.space"
	                frameborder="0"
	                width="1250"
	                height="1450"
                    ></iframe>"""),
                 )

mf_transcribe = gr.Interface(
    fn=transcribe,
    inputs=[
        gr.Microphone(type="filepath"),
        gr.Audio(type="filepath"),
    ],
    outputs="text",
    title=" ",
    description=(
        "Transcribe real-time speech and audio files of any length at the click of a button."
    ),
    allow_flagging="never",
)

yt_transcribe = gr.Interface(
    fn=yt_transcribe,
    inputs=[gr.Textbox(lines=1, placeholder="Paste your YouTube video URL/web address here", label="YouTube Video URL")],
    outputs=["html", "text"],
    title=" ",
    description=(
        "Transcribe YouTube videos at the click of a button."
      
    ),
    allow_flagging="never",
)

clp = gr.TabbedInterface([A1, A5, mf_transcribe, yt_transcribe, A3, A4], ["Chat", "Write", "Transcribe", "Transcribe YouTube Videos", "Describe", "Create"], theme= gr.themes.Glass(primary_hue="neutral", neutral_hue="slate"))
clp.queue().launch()