LLM文档转换 icon

LLM文档转换

LLM文档处理节点,将文档转换为大模型友好的格式

Overview

The node "LLM文档转换" (LLM Document Conversion) provides functionality to convert various document formats into Markdown text using large language model (LLM) processing. Specifically, the "PPT转Markdown" operation under the "文件转Markdown" resource converts PowerPoint presentation files (ppt, pptx) into Markdown format. This is useful for scenarios where users want to extract and transform presentation content into a lightweight, editable Markdown format for documentation, note-taking, or publishing purposes.

Practical examples include:

  • Converting a sales pitch deck into Markdown notes for easy sharing.
  • Transforming training presentations into Markdown-based manuals.
  • Extracting slide content for integration into knowledge bases or wikis.

Properties

Name Meaning
文件字段名 The name of the input file field containing the document to convert. Supported formats include pdf, doc, docx, ppt, pptx, xlsx, html, csv, etc. Default is "data".
返回Markdown文本 Boolean option indicating whether to return the converted Markdown text content. If disabled, only the URL of the converted document is returned. Default is true.

Output

The node outputs JSON data containing the result of the conversion. When "返回Markdown文本" is enabled, the output includes the Markdown text extracted from the PowerPoint file. If disabled, the output provides a URL pointing to the converted Markdown document stored elsewhere.

If binary data is involved (e.g., the original file), it is handled via the specified file field but the main output focuses on the Markdown content or its URL.

Dependencies

  • Requires an API key credential for accessing the LLM document conversion service.
  • The node configuration must include the base URL of the LLM document processing API.
  • Supported document types depend on the backend service capabilities.

Troubleshooting

  • Common issues:

    • Unsupported file format errors if the input file is not one of the supported types.
    • Authentication failures due to missing or invalid API credentials.
    • Network or API endpoint errors if the base URL is incorrect or unreachable.
  • Error messages:

    • "Unsupported file type": Ensure the input file is ppt or pptx for this operation.
    • "Authentication failed": Verify that the API key credential is correctly configured.
    • "Conversion failed": Check the file integrity and retry; also confirm service availability.

Links and References


Note: Internal credential names and service identifiers have been generalized as per instructions.

Discussion