AI Agent for realtime insights on meetings
Workflow JSON
{"nodes": [{"id": "d44489b8-8cb7-4776-8c16-a8bb01e52171", "name": "OpenAI1", "type": "@n8n/n8n-nodes-langchain.openAi", "position": [300, -300], "parameters": {"text": "={{ \n JSON.parse($('Insert Transcription Part').item.json.dialog)\n .filter(item => item.date_updated && new Date(item.date_updated) >= new Date($('Insert Transcription Part').item.json.date_updated))\n .sort((a, b) => a.order - b.order)\n .map(item => `${item.words}\\n${item.speaker}`)\n .join('\\n\\n')\n}}", "memory": "threadId", "prompt": "define", "options": {}, "resource": "assistant", "threadId": "={{ $json.thread_id }}", "assistantId": {"__rl": true, "mode": "list", "value": "asst_D5t6bNnNpenmfC7PmvywMqyR", "cachedResultName": "5minAI - Realtime Agent"}}, "credentials": {"openAiApi": {"id": "", "name": "[Your openAiApi]"}}, "typeVersion": 1.6}, {"id": "3425f1c1-ad68-495e-bb9a-95ea92e7cf23", "name": "Insert Transcription Part", "type": "n8n-nodes-base.postgres", "position": [-120, -300], "parameters": {"query": "UPDATE public.data\nSET output = jsonb_set(\n output,\n '{dialog}', \n (\n COALESCE(\n (output->'dialog')::jsonb, \n '[]'::jsonb -- Initialize as empty array if dialog does not exist\n ) || jsonb_build_object(\n 'order', (COALESCE(jsonb_array_length(output->'dialog'), 0) + 1), -- Calculate the next order\n 'words', '{{ $('Webhook2').item.json.body.data.transcript.words.map(word => word.text.replace(/'/g, \"''\")).join(\" \") }}',\n 'speaker', '{{ $('Webhook2').item.json.body.data.transcript.speaker }}',\n 'language', '{{ $('Webhook2').item.json.body.data.transcript.language }}',\n 'speaker_id', ('{{ $('Webhook2').item.json.body.data.transcript.speaker_id }}')::int,\n 'date_updated', to_jsonb('{{ $now }}'::text)\n )\n )\n)\nWHERE input->>'recall_bot_id' = $1\nReturning input->>'openai_thread_id' as thread_id;", "options": {"queryReplacement": "={{ $('Scenario 2 Start - Webhook').item.json.body.data.bot_id }}"}, "operation": "executeQuery"}, "credentials": {"postgres": {"id": "", "name": "[Your postgres]"}}, "typeVersion": 2.5}, {"id": "9bcc0605-fc35-4842-a3f4-30ef902f35c1", "name": "Create Note", "type": "n8n-nodes-base.postgresTool", "position": [180, -120], "parameters": {"query": "UPDATE public.data\nSET output = jsonb_set(\n output,\n '{notes}', \n (\n COALESCE(\n (output->'notes')::jsonb, \n '[]'::jsonb -- Initialize as empty array if dialog does not exist\n ) || jsonb_build_object(\n 'order', (COALESCE(jsonb_array_length(output->'notes'), 0) + 1), -- Calculate the next order\n 'text', '{{ $fromAI(\"note\",\"Text of note.\") }}'\n )\n )\n)\nWHERE input->>'recall_bot_id' = $1", "options": {"queryReplacement": "={{ $('Scenario 2 Start - Webhook').item.json.body.data.bot_id }}"}, "operation": "executeQuery", "descriptionType": "manual", "toolDescription": "Create note record."}, "credentials": {"postgres": {"id": "", "name": "[Your postgres]"}}, "typeVersion": 2.5}, {"id": "0831c139-ca4b-4b4c-aa7f-7495c4ca0110", "name": "Create Recall bot", "type": "n8n-nodes-base.httpRequest", "position": [-60, -980], "parameters": {"url": "https://us-west-2.recall.ai/api/v1/bot", "method": "POST", "options": {}, "jsonBody": "={\n \"meeting_url\":\"{{ $json.meeting_url }}\",\n \"transcription_options\": {\n \"provider\": \"assembly_ai\"\n }\n,\n\"real_time_transcription\": {\n \"destination_url\": \"https://n8n.lowcoding.dev/webhook/d074ca1e-52f9-47af-8587-8c24d431f9cd\"\n },\n\"automatic_leave\": {\n \"silence_detection\": {\n \"timeout\": 300, \n \"activate_after\": 600\n },\n \"bot_detection\": {\n \"using_participant_events\": {\n \"timeout\": 600, \n \"activate_after\": 1200\n }\n },\n \"waiting_room_timeout\": 600,\n \"noone_joined_timeout\": 600,\n \"everyone_left_timeout\": 2,\n \"in_call_not_recording_timeout\": 600,\n \"recording_permission_denied_timeout\": 600\n}\n}", "sendBody": true, "specifyBody": "json", "authentication": "genericCredentialType", "genericAuthType": "httpHeaderAuth"}, "credentials": {"httpHeaderAuth": {"id": "", "name": "[Your httpHeaderAuth]"}}, "typeVersion": 4.2}, {"id": "e1122b5b-3af5-4836-802c-40c3a0eb3c93", "name": "Create OpenAI thread", "type": "n8n-nodes-base.httpRequest", "position": [140, -980], "parameters": {"url": "https://api.openai.com/v1/threads", "method": "POST", "options": {}, "sendHeaders": true, "authentication": "predefinedCredentialType", "headerParameters": {"parameters": [{"name": "OpenAI-Beta", "value": "assistants=v2"}]}, "nodeCredentialType": "openAiApi"}, "credentials": {"openAiApi": {"id": "", "name": "[Your openAiApi]"}}, "typeVersion": 4.2}, {"id": "784c123d-adbb-4265-9485-2c88dd3091c2", "name": "Create data record", "type": "n8n-nodes-base.supabase", "position": [320, -980], "parameters": {"tableId": "data", "fieldsUi": {"fieldValues": [{"fieldId": "input", "fieldValue": "={{ {\"openai_thread_id\": $('Create OpenAI thread').item.json.id, \"recall_bot_id\": $('Create Recall bot').item.json.id, \"meeting_url\":$('Webhook').item.json.body.meeting_url } }}"}, {"fieldId": "output", "fieldValue": "={{ {\"dialog\":[]} }}"}]}}, "credentials": {"supabaseApi": {"id": "", "name": "[Your supabaseApi]"}}, "typeVersion": 1}, {"id": "f455c7de-1e64-4a28-9eef-11d19c982813", "name": "Sticky Note9", "type": "n8n-nodes-base.stickyNote", "position": [-900, -380], "parameters": {"color": 7, "width": 330.5152611046425, "height": 239.5888196628349, "content": "### ... or watch set up video [10 min]\n[](https://www.youtube.com/watch?v=rtaX6BMiTeo)\n"}, "typeVersion": 1}, {"id": "ea90c110-18ad-4f4b-90ab-fcb88b92e709", "name": "Sticky Note7", "type": "n8n-nodes-base.stickyNote", "position": [-1200, -1060], "parameters": {"color": 7, "width": 636, "height": 657, "content": "\n## AI Agent for realtime insights on meetings\n**Made by [Mark Shcherbakov](https://www.linkedin.com/in/marklowcoding/) from community [5minAI](https://www.skool.com/5minai)**\n\nTranscribing meetings manually can be tedious and prone to error. This workflow automates the transcription process in real-time, ensuring that key discussions and decisions are accurately captured and easily accessible for later review, thus enhancing productivity and clarity in communications.\n\nThe workflow employs an AI-powered assistant to join virtual meetings and capture discussions through real-time transcription. Key functionalities include:\n- Automatic joining of meetings on platforms like Zoom, Google Meet, and others with the ability to provide real-time transcription.\n- Integration with transcription APIs (e.g., AssemblyAI) to deliver seamless and accurate capture of dialogue.\n- Structuring and storing transcriptions efficiently in a database for easy retrieval and analysis.\n\n1. **Real-Time Transcription**: The assistant captures audio during meetings and transcribes it in real-time, allowing participants to focus on discussions.\n2. **Keyword Recognition**: Key phrases can trigger specific actions, such as noting important points or making prompts to the assistant.\n3. **Structured Data Management**: The assistant maintains a database of transcriptions linked to meeting details for organized storage and quick access later."}, "typeVersion": 1}, {"id": "378c19bb-0e4a-43d3-9ba5-2a77ebfb5b83", "name": "Sticky Note6", "type": "n8n-nodes-base.stickyNote", "position": [-1200, -380], "parameters": {"color": 7, "width": 280, "height": 626, "content": "### Set up steps\n\n#### Preparation\n\n1. **Create Recall.ai API key**\n2. **Setup Supabase account and table**\n```\ncreate table\n public.data (\n id uuid not null default gen_random_uuid (),\n date_created timestamp with time zone not null default (now() at time zone 'utc'::text),\n input jsonb null,\n output jsonb null,\n constraint data_pkey primary key (id),\n ) tablespace pg_default;\n\n```\n3. **Create OpenAI API key**\n\n#### Development\n\n1. **Bot Creation**: \n - Use a node to create the bot that will join meetings. Provide the meeting URL and set transcription options within the API request.\n\n2. **Authentication**: \n - Configure authentication settings via a Bearer token for interacting with your transcription service.\n\n3. **Webhook Setup**: \n - Create a webhook to receive real-time transcription updates, ensuring timely data capture during meetings.\n\n4. **Join Meeting**: \n - Set the bot to join the specified meeting and actively listen to capture conversations.\n\n5. **Transcription Handling**: \n - Combine transcription fragments into cohesive sentences and manage dialog arrays for coherence.\n\n6. **Trigger Actions on Keywords**: \n - Set up keyword recognition that can initiate requests to the OpenAI API for additional interactions based on captured dialogue.\n\n7. **Output and Summary Generation**: \n - Produce insights and summary notes from the transcriptions that can be stored back into the database for future reference."}, "typeVersion": 1}, {"id": "9a4ff741-ccfd-42e9-883e-43297a73e2c3", "name": "Scenario 1 Start - Edit Fields", "type": "n8n-nodes-base.set", "position": [-260, -980], "parameters": {"options": {}, "assignments": {"assignments": [{"id": "4891fa6e-2dd5-4433-925c-5497ec82e8ab", "name": "meeting_url", "type": "string", "value": "https://meet.google.com/iix-vrav-kuc"}]}}, "typeVersion": 3.4}, {"id": "a4368763-b96e-45e7-884d-aa0cbae2d276", "name": "Scenario 2 Start - Webhook", "type": "n8n-nodes-base.webhook", "position": [-320, -300], "webhookId": "7f176935-cb83-4147-ac14-48c8d747863a", "parameters": {"path": "d074ca1e-52f9-47af-8587-8c24d431f9cd", "options": {}, "httpMethod": "POST"}, "typeVersion": 2}, {"id": "107b26af-d1d2-40c7-ad4f-7193d3ae9b70", "name": "If Jimmy word", "type": "n8n-nodes-base.if", "position": [80, -300], "parameters": {"options": {}, "conditions": {"options": {"version": 2, "leftValue": "", "caseSensitive": true, "typeValidation": "strict"}, "combinator": "and", "conditions": [{"id": "ba6c2ae5-d0f4-4242-9cf8-97cb84335a93", "operator": {"type": "string", "operation": "contains"}, "leftValue": "={{ $('Scenario 2 Start - Webhook').item.json.body.data.transcript.words.map(word => word.text.replace(/'/g, \"''\")).join(\" \") }}", "rightValue": "=Jimmy"}]}}, "typeVersion": 2.2}, {"id": "49cf34f6-86cf-42cc-9da4-3efb37e6f565", "name": "Sticky Note", "type": "n8n-nodes-base.stickyNote", "position": [-380, -1040], "parameters": {"width": 920, "height": 400, "content": "## Scenario 1\n\n"}, "typeVersion": 1}, {"id": "34660f39-6ecc-4f2d-98e8-a2c529255e98", "name": "Sticky Note1", "type": "n8n-nodes-base.stickyNote", "position": [-380, -360], "parameters": {"width": 1020, "height": 420, "content": "## Scenario 2\n\n"}, "typeVersion": 1}, {"id": "5027e72d-2b2c-40b4-921e-c4f40d85f251", "name": "Sticky Note2", "type": "n8n-nodes-base.stickyNote", "position": [-200, -120], "parameters": {"color": 3, "width": 270, "height": 80, "content": "### Replace Supabase credentials"}, "typeVersion": 1}, {"id": "dddea341-da40-4b6a-ae25-a8417e869cc9", "name": "Sticky Note3", "type": "n8n-nodes-base.stickyNote", "position": [-100, -780], "parameters": {"color": 3, "width": 200, "height": 80, "content": "### Replace server location\n\n"}, "typeVersion": 1}, {"id": "e8e76c2a-f949-400e-92b2-39da8034b471", "name": "Sticky Note4", "type": "n8n-nodes-base.stickyNote", "position": [340, -100], "parameters": {"color": 4, "width": 270, "height": 80, "content": "### Replace OpenAI credentials"}, "typeVersion": 1}, {"id": "729a5f6e-5aea-4908-9a82-2a7d7bea1322", "name": "Sticky Note5", "type": "n8n-nodes-base.stickyNote", "position": [140, -780], "parameters": {"color": 3, "width": 290, "height": 80, "content": "### Replace credentials"}, "typeVersion": 1}, {"id": "31178e90-62ce-4bf8-8381-dc8138088889", "name": "Sticky Note8", "type": "n8n-nodes-base.stickyNote", "position": [-320, -780], "parameters": {"color": 3, "width": 200, "height": 80, "content": "### Replace meeting url\n\n"}, "typeVersion": 1}], "pinData": {"Create Recall bot": [{"id": "ab35fa56-e42b-47c6-b716-eac8d12af601", "join_at": null, "metadata": {}, "recording": null, "video_url": null, "recordings": [], "meeting_url": {"platform": "google_meet", "meeting_id": "zst-ymag-zoa"}, "status_changes": [{"code": "ready", "message": null, "sub_code": null, "created_at": "2024-11-01T11:29:32.364684Z"}], "meeting_metadata": null, "calendar_meetings": [], "meeting_participants": []}], "Insert Transcription Part": [{"dialog": "[{\"order\": 1, \"words\": \"Wait.\", \"speaker\": \"Mark S.\", \"language\": null, \"speaker_id\": 100}, {\"order\": 2, \"words\": \"A bit.\", \"speaker\": \"Mark S.\", \"language\": null, \"speaker_id\": 100}, {\"order\": 3, \"words\": \"It's not even subtitles and it's not even a real. It's. A Google Meet.\", \"speaker\": \"Mark S.\", \"language\": null, \"speaker_id\": 100}, {\"order\": 4, \"words\": \"Same story. I wasn't prepared. I don't know what to tell you. Maybe my AI body can help me.\", \"speaker\": \"Mark S.\", \"language\": null, \"speaker_id\": 100}, {\"order\": 5, \"words\": \"What truth?\", \"speaker\": \"Mark S.\", \"language\": null, \"speaker_id\": 100}, {\"order\": 6, \"words\": \"You can get the same AI body in one day. Just drop AI in comment and I will. Send you a guide.\", \"speaker\": \"Mark S.\", \"language\": null, \"speaker_id\": 100}, {\"order\": 7, \"words\": \"As it works well.\", \"speaker\": \"Mark S.\", \"language\": \"null\", \"speaker_id\": 100}, {\"order\": 8, \"words\": \"As it works well.\", \"speaker\": \"Mark S.\", \"language\": \"null\", \"speaker_id\": 100}, {\"order\": 9, \"words\": \"As it works well.\", \"speaker\": \"Mark S.\", \"language\": \"null\", \"speaker_id\": 100}, {\"order\": 10, \"words\": \"Let's it works well.\", \"speaker\": \"Mark S.\", \"language\": \"null\", \"speaker_id\": 100}, {\"order\": 11, \"words\": \"Let's it works well.\", \"speaker\": \"Mark S.\", \"language\": \"null\", \"speaker_id\": 100}, {\"order\": 12, \"words\": \"Let's it works well.\", \"speaker\": \"Mark S.\", \"language\": \"null\", \"speaker_id\": 100, \"date_updated\": \"2024-11-22T08:41:24.164+01:00\"}, {\"order\": 13, \"words\": \"Let's it works well.\", \"speaker\": \"Mark S.\", \"language\": \"null\", \"speaker_id\": 100, \"date_updated\": \"2024-11-22T08:50:11.330+01:00\"}]", "thread_id": "thread_0g7p3iE7MYmDPiUuPiZP5vfR", "date_updated": "2024-11-22T08:37:55.751+01:00"}], "Scenario 2 Start - Webhook": [{"body": {"data": {"bot_id": "0032c6e2-78e9-46e7-a2ef-41d7b853ef48", "transcript": {"words": [{"text": "Let's", "end_time": 11.88, "start_time": 11.68}, {"text": "it", "end_time": 12.12, "start_time": 11.88}, {"text": "works", "end_time": 12.44, "start_time": 12.12}, {"text": "well.", "end_time": 12.48, "start_time": 12.44}], "source": "smart_annotator", "speaker": "Mark S.", "is_final": true, "language": null, "speaker_id": 100, "original_transcript_id": 32}, "recording_id": "ee1ad589-39fe-4ed5-b96f-cd14c63f3bc2"}, "event": "bot.transcription"}, "query": {}, "params": {}, "headers": {"host": "n8n.lowcoding.dev", "accept": "*/*", "content-type": "application/json", "content-length": "495", "accept-encoding": "gzip", "x-forwarded-for": "52.10.191.34", "x-forwarded-host": "n8n.lowcoding.dev", "x-forwarded-proto": "https"}, "webhookUrl": "https://n8n.lowcoding.dev/webhook/d074ca1e-52f9-47af-8587-8c24d431f9cd", "executionMode": "production"}]}, "connections": {"OpenAI1": {"main": [[]]}, "Create Note": {"ai_tool": [[{"node": "OpenAI1", "type": "ai_tool", "index": 0}]]}, "If Jimmy word": {"main": [[{"node": "OpenAI1", "type": "main", "index": 0}]]}, "Create Recall bot": {"main": [[{"node": "Create OpenAI thread", "type": "main", "index": 0}], []]}, "Create data record": {"main": [[]]}, "Create OpenAI thread": {"main": [[{"node": "Create data record", "type": "main", "index": 0}]]}, "Insert Transcription Part": {"main": [[{"node": "If Jimmy word", "type": "main", "index": 0}]]}, "Scenario 2 Start - Webhook": {"main": [[{"node": "Insert Transcription Part", "type": "main", "index": 0}]]}, "Scenario 1 Start - Edit Fields": {"main": [[{"node": "Create Recall bot", "type": "main", "index": 0}]]}}}How to Import This Workflow
- 1Copy the workflow JSON above using the Copy Workflow JSON button.
- 2Open your n8n instance and go to Workflows.
- 3Click Import from JSON and paste the copied workflow.
Don't have an n8n instance? Start your free trial at n8nautomation.cloud
Related Templates
Text to Speech (OpenAI)
Converts text into natural-sounding speech using OpenAI's Text-to-Speech API. It sends your input text to OpenAI and receives an audio file in return. This is useful for creating audio versions of articles, generating voiceovers for videos, or providing accessibility features for web content. Quickly transform written content into engaging audio.
Automate Customer Support Issue Resolution using AI Text Classifier
Automate the resolution of customer support issues by classifying their state and applying AI-driven actions. This workflow connects Jira for issue management, OpenAI for AI classification and response generation, and Slack for notifications. Support teams can use this to automatically close resolved tickets, remind customers about open issues, or escalate complex cases.
AI-Powered Candidate Shortlisting Automation for ERPNext
Automate AI-powered candidate shortlisting for ERPNext job applications. This workflow connects ERPNext, Google Gemini, WhatsApp, and Outlook to process resumes, evaluate candidates, and communicate outcomes. Recruiters and HR departments can use this to efficiently screen applicants, automatically reject unqualified candidates, and send acceptance notifications. It significantly reduces manual review time and streamlines the hiring process.