AI Meeting Notes → Action Items → Notion

The workflow JSON

Copy or download the full n8n JSON below. Paste it into a new n8n workflow, add your credentials, activate. Full import guide →

{
  "name": "14 - AI Meeting Notes \u2192 Action Items \u2192 Notion",
  "nodes": [
    {
      "parameters": {
        "httpMethod": "POST",
        "path": "meeting-notes",
        "responseMode": "responseNode",
        "options": {
          "binaryData": true
        }
      },
      "id": "node-webhook",
      "name": "Webhook - Audio/Notes Upload",
      "type": "n8n-nodes-base.webhook",
      "typeVersion": 2,
      "position": [
        260,
        300
      ]
    },
    {
      "parameters": {
        "jsCode": "const item = $input.first();\nconst body = item.json.body || item.json;\n\n// Check if this is audio upload (binary) or text transcript\nconst hasBinary = item.binary && Object.keys(item.binary).length > 0;\nconst hasTextTranscript = body.transcript || body.text;\n\nreturn [{\n  json: {\n    meetingTitle: body.title || body.meetingTitle || 'Meeting Notes',\n    meetingDate: body.date || new Date().toISOString().split('T')[0],\n    attendees: body.attendees || body.participants || '',\n    project: body.project || '',\n    submittedBy: body.submittedBy || body.email || '',\n    inputType: hasBinary ? 'audio' : 'text',\n    textTranscript: hasTextTranscript ? (body.transcript || body.text) : null\n  },\n  binary: item.binary || {}\n}];"
      },
      "id": "node-parse-input",
      "name": "Code - Parse Input Type",
      "type": "n8n-nodes-base.code",
      "typeVersion": 2,
      "position": [
        480,
        300
      ]
    },
    {
      "parameters": {
        "conditions": {
          "conditions": [
            {
              "leftValue": "={{ $json.inputType }}",
              "rightValue": "audio",
              "operator": {
                "type": "string",
                "operation": "equals"
              }
            }
          ]
        }
      },
      "id": "node-if-audio",
      "name": "IF - Audio Input?",
      "type": "n8n-nodes-base.if",
      "typeVersion": 2.2,
      "position": [
        700,
        300
      ]
    },
    {
      "parameters": {
        "method": "POST",
        "url": "https://api.openai.com/v1/audio/transcriptions",
        "authentication": "genericCredentialType",
        "genericAuthType": "httpHeaderAuth",
        "sendHeaders": true,
        "headerParameters": {
          "parameters": [
            {
              "name": "Authorization",
              "value": "Bearer {{ $env['OPENAI_API_KEY'] }}"
            }
          ]
        },
        "sendBody": true,
        "contentType": "multipart-form-data",
        "bodyParameters": {
          "parameters": [
            {
              "name": "model",
              "value": "whisper-1"
            },
            {
              "name": "language",
              "value": "en"
            },
            {
              "name": "response_format",
              "value": "verbose_json"
            },
            {
              "name": "timestamp_granularities[]",
              "value": "segment"
            }
          ]
        },
        "options": {
          "bodyContentType": "multipart-form-data"
        }
      },
      "id": "node-whisper",
      "name": "HTTP - Whisper Transcription",
      "type": "n8n-nodes-base.httpRequest",
      "typeVersion": 4.2,
      "position": [
        920,
        180
      ]
    },
    {
      "parameters": {
        "jsCode": "const parsed = $('Code - Parse Input Type').first().json;\nconst whisperResult = $input.first().json;\n\nconst transcript = whisperResult.text || parsed.textTranscript || 'No transcript available';\nconst duration = whisperResult.duration ? `${Math.round(whisperResult.duration / 60)} minutes` : 'Unknown';\n\nreturn [{\n  json: {\n    ...parsed,\n    transcript,\n    duration\n  }\n}];"
      },
      "id": "node-merge-transcript",
      "name": "Code - Merge Transcript",
      "type": "n8n-nodes-base.code",
      "typeVersion": 2,
      "position": [
        1140,
        300
      ]
    },
    {
      "parameters": {
        "resource": "text",
        "operation": "message",
        "modelId": {
          "__rl": true,
          "value": "gpt-4o",
          "mode": "list"
        },
        "messages": {
          "values": [
            {
              "role": "system",
              "content": "You are an expert meeting facilitator and note-taker. Analyze the meeting transcript and produce structured output.\n\nReturn a JSON object with:\n{\n  \"summary\": \"2-3 paragraph executive summary of the meeting\",\n  \"key_decisions\": [\"Decision 1\", \"Decision 2\"],\n  \"action_items\": [\n    {\n      \"task\": \"Clear task description\",\n      \"owner\": \"Person's name (or 'TBD')\",\n      \"due_date\": \"YYYY-MM-DD or 'No date set'\",\n      \"priority\": \"High|Medium|Low\"\n    }\n  ],\n  \"blockers\": [\"Blocker 1\"],\n  \"topics_discussed\": [\"Topic 1\", \"Topic 2\"],\n  \"next_meeting\": \"Date/time if mentioned, or 'Not scheduled'\"\n}\n\nBe specific and actionable. Extract real names from the transcript for task owners."
            },
            {
              "role": "user",
              "content": "Meeting: {{ $json.meetingTitle }}\nDate: {{ $json.meetingDate }}\nAttendees: {{ $json.attendees }}\nProject: {{ $json.project }}\nDuration: {{ $json.duration }}\n\nTranscript:\n{{ $json.transcript }}"
            }
          ]
        },
        "options": {
          "temperature": 0.3,
          "maxTokens": 2000
        }
      },
      "id": "node-gpt-analyze",
      "name": "GPT-4o - Analyze Meeting",
      "type": "@n8n/n8n-nodes-langchain.openAi",
      "typeVersion": 1.7,
      "position": [
        1360,
        300
      ],
      "credentials": {
        "openAiApi": {
          "name": "<your credential>"
        }
      }
    },
    {
      "parameters": {
        "jsCode": "const meetingData = $('Code - Merge Transcript').first().json;\nconst aiRaw = $input.first().json.message?.content || '{}';\n\nlet analysis;\ntry { analysis = JSON.parse(aiRaw); }\ncatch (e) { analysis = { summary: aiRaw, key_decisions: [], action_items: [], blockers: [], topics_discussed: [], next_meeting: 'Not scheduled' }; }\n\n// Build Notion page blocks\nconst actionItemsBlocks = (analysis.action_items || []).map(item =>\n  `\u2610 **${item.task}** \u2014 Owner: ${item.owner} | Due: ${item.due_date} | Priority: ${item.priority}`\n).join('\\n');\n\nconst decisionsText = (analysis.key_decisions || []).map(d => `\u2022 ${d}`).join('\\n');\nconst blockersText = (analysis.blockers || []).map(b => `\u26a0\ufe0f ${b}`).join('\\n');\nconst topicsText = (analysis.topics_discussed || []).map(t => `\u2022 ${t}`).join('\\n');\n\nreturn [{\n  json: {\n    ...meetingData,\n    analysis,\n    actionItemsBlocks,\n    decisionsText,\n    blockersText,\n    topicsText,\n    pageTitle: `${meetingData.meetingDate} \u2014 ${meetingData.meetingTitle}`\n  }\n}];"
      },
      "id": "node-parse-analysis",
      "name": "Code - Parse & Format Analysis",
      "type": "n8n-nodes-base.code",
      "typeVersion": 2,
      "position": [
        1580,
        300
      ]
    },
    {
      "parameters": {
        "resource": "page",
        "operation": "create",
        "databaseId": {
          "__rl": true,
          "value": "YOUR_NOTION_MEETINGS_DB",
          "mode": "id"
        },
        "title": "={{ $json.pageTitle }}",
        "propertiesUi": {
          "propertyValues": [
            {
              "key": "Date",
              "type": "date",
              "date": "={{ $json.meetingDate }}"
            },
            {
              "key": "Project",
              "type": "rich_text",
              "textContent": "={{ $json.project }}"
            },
            {
              "key": "Attendees",
              "type": "rich_text",
              "textContent": "={{ $json.attendees }}"
            },
            {
              "key": "Duration",
              "type": "rich_text",
              "textContent": "={{ $json.duration }}"
            },
            {
              "key": "Action Items Count",
              "type": "number",
              "numberValue": "={{ ($json.analysis.action_items || []).length }}"
            },
            {
              "key": "Next Meeting",
              "type": "rich_text",
              "textContent": "={{ $json.analysis.next_meeting }}"
            }
          ]
        },
        "blockUi": {
          "blockValues": [
            {
              "type": "heading_2",
              "textContent": "\ud83d\udccb Summary"
            },
            {
              "type": "paragraph",
              "textContent": "={{ $json.analysis.summary }}"
            },
            {
              "type": "divider"
            },
            {
              "type": "heading_2",
              "textContent": "\u2705 Action Items"
            },
            {
              "type": "paragraph",
              "textContent": "={{ $json.actionItemsBlocks }}"
            },
            {
              "type": "divider"
            },
            {
              "type": "heading_2",
              "textContent": "\ud83c\udfaf Key Decisions"
            },
            {
              "type": "paragraph",
              "textContent": "={{ $json.decisionsText }}"
            },
            {
              "type": "heading_2",
              "textContent": "\ud83d\udd35 Topics Discussed"
            },
            {
              "type": "paragraph",
              "textContent": "={{ $json.topicsText }}"
            },
            {
              "type": "heading_2",
              "textContent": "\u26a0\ufe0f Blockers & Risks"
            },
            {
              "type": "paragraph",
              "textContent": "={{ $json.blockersText || 'None identified' }}"
            },
            {
              "type": "divider"
            },
            {
              "type": "heading_2",
              "textContent": "\ud83d\udcdd Full Transcript"
            },
            {
              "type": "paragraph",
              "textContent": "={{ $json.transcript }}"
            }
          ]
        }
      },
      "id": "node-notion-create",
      "name": "Notion - Create Meeting Page",
      "type": "n8n-nodes-base.notion",
      "typeVersion": 2.2,
      "position": [
        1800,
        300
      ],
      "credentials": {
        "notionApi": {
          "name": "<your credential>"
        }
      }
    },
    {
      "parameters": {
        "respondWith": "json",
        "responseBody": "={\n  \"success\": true,\n  \"notionUrl\": \"{{ $json.url }}\",\n  \"actionItems\": {{ $('Code - Parse & Format Analysis').item.json.analysis.action_items?.length || 0 }},\n  \"message\": \"Meeting notes processed and saved to Notion\"\n}"
      },
      "id": "node-respond",
      "name": "Respond to Webhook",
      "type": "n8n-nodes-base.respondToWebhook",
      "typeVersion": 1,
      "position": [
        2020,
        300
      ]
    }
  ],
  "connections": {
    "Webhook - Audio/Notes Upload": {
      "main": [
        [
          {
            "node": "Code - Parse Input Type",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Code - Parse Input Type": {
      "main": [
        [
          {
            "node": "IF - Audio Input?",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "IF - Audio Input?": {
      "main": [
        [
          {
            "node": "HTTP - Whisper Transcription",
            "type": "main",
            "index": 0
          }
        ],
        [
          {
            "node": "Code - Merge Transcript",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "HTTP - Whisper Transcription": {
      "main": [
        [
          {
            "node": "Code - Merge Transcript",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Code - Merge Transcript": {
      "main": [
        [
          {
            "node": "GPT-4o - Analyze Meeting",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "GPT-4o - Analyze Meeting": {
      "main": [
        [
          {
            "node": "Code - Parse & Format Analysis",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Code - Parse & Format Analysis": {
      "main": [
        [
          {
            "node": "Notion - Create Meeting Page",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Notion - Create Meeting Page": {
      "main": [
        [
          {
            "node": "Respond to Webhook",
            "type": "main",
            "index": 0
          }
        ]
      ]
    }
  },
  "active": false,
  "settings": {
    "executionOrder": "v1"
  },
  "tags": [
    {
      "name": "meetings"
    },
    {
      "name": "ai"
    },
    {
      "name": "notion"
    },
    {
      "name": "productivity"
    }
  ],
  "meta": {
    "description": "Upload audio recording or text transcript via webhook \u2192 Whisper transcribes audio \u2192 GPT-4o extracts summary/action items/decisions \u2192 creates structured Notion page with full transcript and formatted action items.",
    "prerequisites": [
      "OpenAI API key (Whisper + GPT-4o)",
      "Notion integration with meetings database access",
      "Notion database with: Date, Project, Attendees, Duration, Action Items Count (number), Next Meeting",
      "Set OPENAI_API_KEY env variable",
      "Audio formats supported by Whisper: mp3, mp4, wav, m4a, webm (max 25MB)"
    ],
    "testingScenario": {
      "text_test_payload": {
        "title": "Q2 Planning",
        "date": "2026-06-13",
        "attendees": "Santoshi, Alex, Jordan",
        "project": "The AI Stack",
        "transcript": "Santoshi: Let's discuss the Q2 roadmap. Alex: I think we need to prioritize the newsletter automation. Jordan: Agreed. I'll own that by June 20th. Santoshi: Great. Let's also fix the lead routing bug. Jordan can handle that too, due June 17th. Any blockers? Alex: We're waiting on API keys from Beehiiv. That's blocking the newsletter work. Next meeting: June 20th 10 AM."
      }
    }
  }
}

Credentials you'll need

Each integration node will prompt for credentials when you import. We strip credential IDs before publishing — you'll add your own.

• notionApi
• openAiApi

About this workflow

14 - AI Meeting Notes → Action Items → Notion. Uses httpRequest, openAi, notion. Webhook trigger; 9 nodes.

Source: https://github.com/satmakuru222/TheAIStackk/blob/main/n8n-workflows/14-ai-meeting-notes-action-items-notion.json — original creator credit. Request a take-down →

More AI & RAG workflows → · Browse all categories →

Related workflows

Workflows that share integrations, category, or trigger type with this one. All free to copy and import.