Automated Project Summarization from Meeting Transcripts using GPT-4 and Google Docs - n8n Workflow

Project Managers and Consultants needing standardized client documentation.
Automation Specialists looking to deploy advanced AI workflows.
Teams handling large volumes of transcribed meeting data.
Users seeking robust, pre-built n8n templates for language model applications.

Generating professional project summary documents from lengthy meeting transcripts is often a tedious, manual task. This robust n8n workflow solves this by fully automating the process using advanced large language models (LLMs).

When executed, the automation pulls a raw transcript from Google Docs, passes it to the GPT-4 model with a highly detailed, professional prompt structure (including Project Overview, Key Challenges, Deliverables, and Timeline). The resulting output is a clean, structured Markdown document.

Crucially, this n8n automation template then handles the conversion and secure saving process, utilizing a Code n8n node and HTTP Request n8n node to correctly format and upload the summary as a native Google Doc file to your desired Google Drive folder, ensuring professional formatting is maintained. This saves significant time and guarantees consistency across all project documentation, showcasing the power of complex n8n workflows.

This powerful automation starts with the manual trigger n8n node, allowing the user to initiate the process on demand.

1. Trigger & Retrieval: The process begins when the user clicks 'Execute Workflow' on the Manual Trigger n8n node. It immediately proceeds to the Google Docs n8n node (Get meeting transcript) to retrieve the content of a specified meeting transcript document.


2. AI Processing: The retrieved content is routed to the Langchain Chain n8n node (Project Summary). This n8n node uses an extensive system prompt, positioning the AI as a professional project manager tasked with creating a structured summary, complete with specific headings and a required timeline format. This chain is powered by the OpenAI Chat Model n8n node, utilizing a GPT-4 version for high-quality reasoning.


3. Formatting Conversion: The AI returns the summary in Markdown format. The next n8n node, Markdown, converts this output into correctly structured HTML content, preparing it for Google Docs.


4. Configuration: A Set n8n node (setfields) defines the name of the new document ("Project Summary") and the target Google Drive folder ID.


5. Request Preparation: The critical Code n8n node (PrepareRequest) dynamically constructs the required multipart/related request body, including file metadata (name, MIME type for Google Docs) and the styled HTML content. This custom scripting is essential for successfully creating a native Google Doc via the API.


6. Document Creation: Finally, the HTTP Request n8n node (CreateGoogleDoc) executes the request against the Google Drive API's upload endpoint, saving the newly generated, structured project summary document to the specified location. The entire n8n workflow concludes with a No Operation n8n node.

To deploy this expert n8n workflow template, follow these steps:

1. Import the n8n Workflow: Copy the provided JSON code and paste it into your n8n instance using the 'New' -> 'Import from JSON' option.


2. Configure Credentials: You will need three credentials set up for this n8n template to function:

1. Test: Execute the n8n trigger manually to ensure all steps run correctly and the summary document is created.

This n8n workflow leverages specialized and core n8n node types:

Manual Trigger n8n node: (When clicking ‘Execute workflow’)
Function: Starts the n8n workflow execution upon user command.
Key Configuration: Simple manual execution setup.
Google Docs n8n node: (Get meeting transcript)
Function: Connects to Google Docs to retrieve the raw text content of the source meeting transcript.
Key Configuration: Uses the get operation and requires the specific document URL/ID.
Langchain Chain LLM n8n node: (Project Summary)
Function: Serves as the prompt orchestrator, passing the transcript and the detailed structure template to the LLM.
Key Configuration: Contains a highly detailed prompt ensuring structured output (markdown format).
OpenAI Chat Model n8n node: (OpenAI Chat Model)
Function: Executes the AI request using the gpt-4.1-mini model (via Langchain) to generate the summary.
Key Configuration: Utilizes high timeout (300000 ms) and connects directly to the Langchain chain.
Markdown n8n node: (Markdown)
Function: Converts the Markdown output from the AI into HTML, which is a required format for uploading styled content to Google Docs via API.
Key Configuration: Mode set to markdownToHtml.
Code n8n node: (Prepare_Request)
Function: Custom code to construct the complex multipart/related API payload necessary for creating a native Google Doc with embedded HTML styling and metadata.
Key Configuration: Defines boundaries, file name, folder ID, and injects custom CSS for clean document spacing.
HTTP Request n8n node: (CreateGoogleDoc)
Function: Performs the final API call to Google Drive's upload endpoint to create the new summary file.
* Key Configuration: Method POST, Content-Type raw (multipart/related), and uses a Google Drive OAuth credential.