{ "cells": [ { "cell_type": "markdown", "id": "d12b9c22", "metadata": {}, "source": [ "# Song Lyrics → One-Sentence Summary\n", "Get the lyrics of a song and summarize its main idea in about one sentence.\n", "\n", "## Setup\n", "Import required libraries: environment vars, display helper, OpenAI client, BeautifulSoup, and requests." ] }, { "cell_type": "code", "execution_count": null, "id": "d94bbd61", "metadata": {}, "outputs": [], "source": [ "import os\n", "from IPython.display import Markdown, display\n", "from openai import OpenAI\n", "from bs4 import BeautifulSoup\n", "import requests" ] }, { "cell_type": "markdown", "id": "92dc1bde", "metadata": {}, "source": [ "## Function: Get Lyrics from Genius\n", "Fetch and extract the lyrics from a Genius.com song page using BeautifulSoup." ] }, { "cell_type": "code", "execution_count": null, "id": "2b43fa98", "metadata": {}, "outputs": [], "source": [ "def get_lyrics_from_genius(url: str) -> str:\n", " \"\"\"\n", " Extracts song lyrics from a Genius.com song URL using BeautifulSoup.\n", " Example URL: https://genius.com/Ed-sheeran-shape-of-you-lyrics\n", " \"\"\"\n", " # Standard headers to fetch a website\n", " headers = {\n", " \"User-Agent\": \"Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/117.0.0.0 Safari/537.36\"\n", " }\n", "\n", " response = requests.get(url, headers=headers)\n", " response.raise_for_status() # raises error if page not found\n", "\n", " soup = BeautifulSoup(response.text, \"html.parser\")\n", "\n", " # Genius stores lyrics inside
\n", " lyrics_blocks = soup.find_all(\"div\", {\"data-lyrics-container\": \"true\"})\n", "\n", " if not lyrics_blocks:\n", " return \"Lyrics not found.\"\n", "\n", " # Join all text blocks and clean up spacing\n", " lyrics = \"\\n\".join(block.get_text(separator=\"\\n\") for block in lyrics_blocks)\n", " return lyrics.strip()" ] }, { "cell_type": "markdown", "id": "fc4f0590", "metadata": {}, "source": [ "## Function: Create Genius URL\n", "Build a Genius.com lyrics URL automatically from the given artist and song name." ] }, { "cell_type": "code", "execution_count": null, "id": "e018c623", "metadata": {}, "outputs": [], "source": [ "def create_genius_url(artist: str, song: str) -> str:\n", " \"\"\"\n", " Creates a Genius.com lyrics URL from artist and song name.\n", " Example:\n", " create_genius_url(\"Ed sheeran\", \"shape of you\")\n", " → https://genius.com/Ed-sheeran-shape-of-you-lyrics\n", " \"\"\"\n", " artist = artist.strip().replace(\" \", \"-\")\n", " song = song.strip().replace(\" \", \"-\")\n", " return f\"https://genius.com/{artist}-{song}-lyrics\"\n" ] }, { "cell_type": "markdown", "id": "62f50f02", "metadata": {}, "source": [ "## Generate URL and Fetch Lyrics\n", "Create the Genius URL from the artist and song name, then fetch and display the lyrics." ] }, { "cell_type": "code", "execution_count": null, "id": "ed51d48d", "metadata": {}, "outputs": [], "source": [ "artist = \"Ed sheeran\"\n", "song = \"shape of you\"\n", "\n", "url = create_genius_url(artist, song)\n", "print(url)\n", "# Output: https://genius.com/Ed-sheeran-shape-of-you-lyrics\n", "\n", "user_prompt = get_lyrics_from_genius(url)\n", "print(user_prompt[:5000]) " ] }, { "cell_type": "code", "execution_count": null, "id": "fca4203a", "metadata": {}, "outputs": [], "source": [ "system_prompt = \"\"\"\n", "You are a **helpful assistant** that specializes in analyzing **song lyrics**.\n", "\n", "## Task\n", "Your goal is to **summarize the main idea or theme of a song** in **about one sentence**.\n", "\n", "## Instructions\n", "1. Read the given song lyrics carefully.\n", "2. Identify the **core message**, **emotion**, or **story** of the song.\n", "3. Respond with **one concise sentence** only.\n", "4. The tone of your summary should reflect the song’s mood (e.g., joyful, melancholic, romantic, rebellious).\n", "\n", "## Edge Cases\n", "- **Very short lyrics:** Summarize the implied meaning.\n", "- **Repetitive lyrics:** Focus on the message or emotion being emphasized.\n", "- **Abstract or nonsensical lyrics:** Describe the overall feeling or imagery they create.\n", "- **No lyrics or only a title provided:** Reply with \n", " `No lyrics provided — unable to summarize meaningfully.`\n", "- **Non-English lyrics:** Summarize in English unless otherwise instructed.\n", "\n", "## Output Format\n", "Plain text — a single, coherent sentence summarizing the main idea of the song.\n", "\"\"\"" ] }, { "cell_type": "markdown", "id": "11784d62", "metadata": {}, "source": [ "## Create Chat Messages\n", "Prepare the system and user messages, then send them to the OpenAI model for summarization." ] }, { "cell_type": "code", "execution_count": null, "id": "f1205658", "metadata": {}, "outputs": [], "source": [ "messages = [\n", " {\"role\": \"system\", \"content\": system_prompt},\n", " {\"role\": \"user\", \"content\": user_prompt}\n", "]" ] }, { "cell_type": "code", "execution_count": null, "id": "5c8d61aa", "metadata": {}, "outputs": [], "source": [ "openai = OpenAI()\n", "response = openai.chat.completions.create(\n", " model = \"gpt-4.1-mini\",\n", " messages = messages\n", ")" ] }, { "cell_type": "markdown", "id": "4ad95820", "metadata": {}, "source": [ "## Display Summary\n", "Show the model’s one-sentence summary of the song lyrics in a formatted Markdown output." ] }, { "cell_type": "code", "execution_count": null, "id": "4f09a642", "metadata": {}, "outputs": [], "source": [ "display(Markdown(response.choices[0].message.content))" ] } ], "metadata": { "kernelspec": { "display_name": ".venv", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.12.6" } }, "nbformat": 4, "nbformat_minor": 5 }