app.py

import streamlit as st
from audiorecorder import audiorecorder
import os
import openai
import requests
import re
def create_recipe(path):
    import whisper
    # small model for transcription
    model = whisper.load_model("base")
    result = model.transcribe("audio.wav")
    # print(result["text"])
    openai.api_type = "azure"
    openai.api_version = "2023-05-15"
    openai.api_base = "https://futurice-data-day-2023.openai.azure.com/"
    openai.api_key = "d2cb77316cee4feb9d96d70ed77ef27d"
    response = openai.ChatCompletion.create(
        engine="gpt-35-16k",  # engine options = ["gpt-35-16k", "gpt-4", "gpt-4-32k"]
        messages=[
            {"role": "system", "content": "You are a Recipe generator."},
            {"role": "user", "content": result["text"]},
            {"role": "assistant", "content": "Create a recipe based on user's dietary restriction and personalize it. Write the recipe name after word 'Recipe:' and then give the ingredients after word 'Ingredients:' and instruction after word 'Instruction:'"},
        ],
    )
    # print(response)
    # print(response["choices"][0]["message"]["content"])
    recipe_text = response["choices"][0]["message"]["content"]
    return recipe_text
def create_recipe_image(recipe_text):
    openai.api_base = "https://futurice-data-day-2023.openai.azure.com/"
    openai.api_key = "d2cb77316cee4feb9d96d70ed77ef27d"
    # Assign the API version (DALL-E is currently supported for the 2023-06-01-preview API version only)
    openai.api_version = "2023-06-01-preview"
    openai.api_type = "azure"
    pattern = re.compile(r'Recipe:(.*?)Ingredients:', re.DOTALL)
    # Use re.search() to find the matching portion of the text.
    match = pattern.search(recipe_text)
    if match:
        extracted_text = match.group(1).strip()
    generation_response = openai.Image.create(
    prompt=extracted_text, size="1024x1024", n=2  # Enter your prompt text here
    )
    # Set the directory for the stored image
    image_dir = os.path.join(os.curdir, "images")
    # If the directory doesn't exist, create it
    if not os.path.isdir(image_dir):
        os.mkdir(image_dir)
    # Initialize the image path (note the filetype should be png)
    image_path = os.path.join(image_dir, "generated_image.png")
    # # Retrieve the generated image
    image_url = generation_response["data"][0]["url"]  # extract image URL from response
    generated_image = requests.get(image_url).content  # download the image
    with open(image_path, "wb") as image_file:
        image_file.write(generated_image)
    from PIL import Image
    image = Image.open(image_path)
    st.image(image, caption='Recipe')
    st.write(recipe_text)
st.markdown(""" <style> .font {
font-size:50px ; font-family: 'Cooper Black'; color: #FF9633;}
</style> """, unsafe_allow_html=True)
st.markdown('<p class="font">Cooking with a Dash of AI: Recipe Generator Delivers Delicious Delights!</p>', unsafe_allow_html=True)
# st.title("Cooking with a Dash of AI: Recipe Generator Delivers Delicious Delights!")
st.markdown(""" <style> .font2 {
font-size:30px ; font-family: 'Cooper Black'; color: #000000;}
</style> """, unsafe_allow_html=True)
st.markdown('<p class="font2">Could you please share your favorite dish and any dietary constraints you have?</p>', unsafe_allow_html=True)
audio = audiorecorder("Click to record your voice", "Click to stop recording")
if len(audio) > 0:
    # To play audio in frontend:
    # st.audio(audio.export().read())
    # # To save audio to a file, use pydub export method:
    audio.export("audio.wav", format="wav")
    recipe_text = create_recipe("audio.wav")
    create_recipe_image(recipe_text)
    # To get audio properties, use pydub AudioSegment properties:
    # st.write(f"Frame rate: {audio.frame_rate}, Frame width: {audio.frame_width}, Duration: {audio.duration_seconds} seconds")