app.py

import streamlit as st
from openai import OpenAI
import os
import base64
import cv2
from moviepy.editor import VideoFileClip

API_KEY = os.getenv('gpt4okey')

# Set the API key and model name
MODEL = "gpt-4o"

# Switch to project based with limits and use org id and key to identify run pool
# models for GPT-4o project:    gpt-4o-2024-05-13   (gpt-4o)
#client = OpenAI(api_key=os.environ.get("OPENAI_API_KEY", "<your OpenAI API key if not set as an env var>"))
client = OpenAI(api_key=API_KEY)


def process_text():
    text_input = st.text_input("Enter your text:")
    if text_input:
        completion = client.chat.completions.create(
            model=MODEL,
            messages=[
                {"role": "system", "content": "You are a helpful assistant. Help me with my math homework!"},
                {"role": "user", "content": f"Hello! Could you solve {text_input}?"}
            ]
        )
        st.write("Assistant: " + completion.choices[0].message.content)

def process_image(image_input):
    if image_input:
        base64_image = base64.b64encode(image_input.read()).decode("utf-8")
        response = client.chat.completions.create(
            model=MODEL,
            messages=[
                {"role": "system", "content": "You are a helpful assistant that responds in Markdown. Help me with my math homework!"},
                {"role": "user", "content": [
                    {"type": "text", "text": "What's the area of the triangle?"},
                    {"type": "image_url", "image_url": {
                        "url": f"data:image/png;base64,{base64_image}"}
                    }
                ]}
            ],
            temperature=0.0,
        )
        st.markdown(response.choices[0].message.content)

def process_audio(audio_input):
    if audio_input:
        transcription = client.audio.transcriptions.create(
            model="whisper-1",
            file=audio_input,
        )
        response = client.chat.completions.create(
            model=MODEL,
            messages=[
                {"role": "system", "content": "You are generating a transcript summary. Create a summary of the provided transcription. Respond in Markdown."},
                {"role": "user", "content": [
                    {"type": "text", "text": f"The audio transcription is: {transcription.text}"}
                ]},
            ],
            temperature=0,
        )
        st.markdown(response.choices[0].message.content)

def process_video(video_input):
    if video_input:
        base64Frames, audio_path = process_video_frames(video_input)
        transcription = client.audio.transcriptions.create(
            model="whisper-1",
            file=open(audio_path, "rb"),
        )
        response = client.chat.completions.create(
            model=MODEL,
            messages=[
                {"role": "system", "content": "You are generating a video summary. Create a summary of the provided video and its transcript. Respond in Markdown"},
                {"role": "user", "content": [
                    "These are the frames from the video.",
                    *map(lambda x: {"type": "image_url",
                                    "image_url": {"url": f'data:image/jpg;base64,{x}', "detail": "low"}}, base64Frames),
                    {"type": "text", "text": f"The audio transcription is: {transcription.text}"}
                ]},
            ],
            temperature=0,
        )
        st.markdown(response.choices[0].message.content)

def process_video_frames(video_path, seconds_per_frame=2):
    base64Frames = []
    base_video_path, _ = os.path.splitext(video_path.name)
    video = cv2.VideoCapture(video_path.name)
    total_frames = int(video.get(cv2.CAP_PROP_FRAME_COUNT))
    fps = video.get(cv2.CAP_PROP_FPS)
    frames_to_skip = int(fps * seconds_per_frame)
    curr_frame = 0
    while curr_frame < total_frames - 1:
        video.set(cv2.CAP_PROP_POS_FRAMES, curr_frame)
        success, frame = video.read()
        if not success:
            break
        _, buffer = cv2.imencode(".jpg", frame)
        base64Frames.append(base64.b64encode(buffer).decode("utf-8"))
        curr_frame += frames_to_skip
    video.release()
    audio_path = f"{base_video_path}.mp3"
    clip = VideoFileClip(video_path.name)
    clip.audio.write_audiofile(audio_path, bitrate="32k")
    clip.audio.close()
    clip.close()
    return base64Frames, audio_path

def main():
    st.title("Omni Demo")
    option = st.selectbox("Select an option", ("Text", "Image", "Audio", "Video"))
    if option == "Text":
        process_text()
    elif option == "Image":
        image_input = st.file_uploader("Upload an image", type=["jpg", "jpeg", "png"])
        process_image(image_input)
    elif option == "Audio":
        audio_input = st.file_uploader("Upload an audio file", type=["mp3", "wav"])
        process_audio(audio_input)
    elif option == "Video":
        video_input = st.file_uploader("Upload a video file", type=["mp4"])
        process_video(video_input)

if __name__ == "__main__":
    main()