cognee/notebooks/cognee_multimedia_demo.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Cognee GraphRAG with Multimedia files"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "vscode": {
     "languageId": "plaintext"
    }
   },
   "source": [
    "## Load Data\n",
    "\n",
    "We will use a few sample multimedia files which we have on GitHub for easy access."
   ]
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-06-30T11:54:44.613431Z",
     "start_time": "2025-06-30T11:54:44.606687Z"
    }
   },
   "cell_type": "code",
   "source": [
    "import os\n",
    "import pathlib\n",
    "\n",
    "# cognee knowledge graph will be created based on the text\n",
    "# and description of these files\n",
    "mp3_file_path = os.path.join(\n",
    "    os.path.abspath(\"\"),\n",
    "    \"../\",\n",
    "    \"examples/data/multimedia/text_to_speech.mp3\",\n",
    ")\n",
    "png_file_path = os.path.join(\n",
    "    os.path.abspath(\"\"),\n",
    "    \"../\",\n",
    "    \"examples/data/multimedia/example.png\",\n",
    ")"
   ],
   "outputs": [],
   "execution_count": 1
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Set environment variables"
   ]
  },
  {
   "cell_type": "code",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-06-30T11:54:46.739157Z",
     "start_time": "2025-06-30T11:54:46.734808Z"
    }
   },
   "source": [
    "import os\n",
    "\n",
    "if \"LLM_API_KEY\" not in os.environ:\n",
    "    os.environ[\"LLM_API_KEY\"] = \"\"\n",
    "\n",
    "# \"neo4j\" or \"networkx\"\n",
    "os.environ[\"GRAPH_DATABASE_PROVIDER\"] = \"kuzu\"\n",
    "# Not needed if using networkx\n",
    "# os.environ[\"GRAPH_DATABASE_URL\"]=\"\"\n",
    "# os.environ[\"GRAPH_DATABASE_USERNAME\"]=\"\"\n",
    "# os.environ[\"GRAPH_DATABASE_PASSWORD\"]=\"\"\n",
    "\n",
    "# \"pgvector\", \"qdrant\", \"weaviate\" or \"lancedb\"\n",
    "os.environ[\"VECTOR_DB_PROVIDER\"] = \"lancedb\"\n",
    "# Not needed if using \"lancedb\" or \"pgvector\"\n",
    "# os.environ[\"VECTOR_DB_URL\"]=\"\"\n",
    "# os.environ[\"VECTOR_DB_KEY\"]=\"\"\n",
    "\n",
    "# Relational Database provider \"sqlite\" or \"postgres\"\n",
    "os.environ[\"DB_PROVIDER\"] = \"sqlite\"\n",
    "\n",
    "# Database name\n",
    "os.environ[\"DB_NAME\"] = \"cognee_db\"\n",
    "\n",
    "# Postgres specific parameters (Only if Postgres or PGVector is used)\n",
    "# os.environ[\"DB_HOST\"]=\"127.0.0.1\"\n",
    "# os.environ[\"DB_PORT\"]=\"5432\"\n",
    "# os.environ[\"DB_USERNAME\"]=\"cognee\"\n",
    "# os.environ[\"DB_PASSWORD\"]=\"cognee\""
   ],
   "outputs": [],
   "execution_count": 2
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Run Cognee with multimedia files"
   ]
  },
  {
   "cell_type": "code",
   "metadata": {},
   "source": [
    "import cognee\n",
    "\n",
    "# Create a clean slate for cognee -- reset data and system state\n",
    "await cognee.prune.prune_data()\n",
    "await cognee.prune.prune_system(metadata=True)\n",
    "\n",
    "# Add multimedia files and make them available for cognify\n",
    "await cognee.add([mp3_file_path, png_file_path])\n",
    "\n",
    "# Create knowledge graph with cognee\n",
    "await cognee.cognify()"
   ],
   "outputs": [],
   "execution_count": null
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Query Cognee for summaries related to multimedia files"
   ]
  },
  {
   "cell_type": "code",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2025-06-30T11:44:56.372628Z",
     "start_time": "2025-06-30T11:44:55.978258Z"
    }
   },
   "source": [
    "from cognee.api.v1.search import SearchType\n",
    "\n",
    "# Query cognee for summaries of the data in the multimedia files\n",
    "search_results = await cognee.search(\n",
    "    query_type=SearchType.SUMMARIES,\n",
    "    query_text=\"What is in the multimedia files?\",\n",
    ")\n",
    "\n",
    "# Display search results\n",
    "for result_text in search_results:\n",
    "    print(result_text)"
   ],
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'id': '3b530220-7e7c-52a2-8b62-ce5adce1a46c', 'created_at': 1751283883122, 'updated_at': 1751283883122, 'ontology_valid': False, 'version': 1, 'topological_rank': 0, 'type': 'IndexSchema', 'text': \"The joke queries the number of programmers required to change a light bulb and answers, 'None. That’s a hardware issue.' This humor highlights the divide between software and hardware challenges in programming.\"}\n",
      "{'id': '128eb96e-fd36-53ef-ab6d-d4884ecbfee9', 'created_at': 1751283883122, 'updated_at': 1751283883122, 'ontology_valid': False, 'version': 1, 'topological_rank': 0, 'type': 'IndexSchema', 'text': \"Changing a light bulb doesn't require programmers.\"}\n"
     ]
    }
   ],
   "execution_count": 5
  }
 ],
 "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": "3.9.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}