Actions21
Overview
This node provides document and image parsing capabilities using the 阿里通义 (Ali Tongyi) service. It is designed to analyze images by either URL or binary data input, returning parsed textual information or other extracted content from the images. This functionality is useful in scenarios such as:
- Automatically extracting text or metadata from images for further processing.
- Analyzing multiple images at once by providing multiple URLs.
- Integrating image understanding into workflows that require AI-based image recognition or OCR.
For example, a user might provide an image URL of a product label and receive back the recognized text or description, enabling automated cataloging or compliance checks.
Properties
| Name | Meaning |
|---|---|
| 模型 (assistantId) | The model or intelligent agent used for processing. Can be any string; if unknown, can be arbitrary. |
| 文本输入 (text) | Text input related to the image, e.g., a question like "What is this image?" |
| 输入类型 (inputType) | Type of input provided: either "图片链接" (image URL) or "二进制文件" (binary file in base64 format) |
| URL链接 (imageUrls) | One or more image URLs to analyze, separated by commas. Required if inputType is "url". |
| 输入数据字段名称 (binaryPropertyName) | Name of the binary property field containing the image data when inputType is "base64". Default is "data". |
| 简化输出 (simplify) | Whether to simplify the response output. Boolean value. Defaults to true. |
| 语音列表 (voice) | Voice selection for text-to-speech features, with options including official voices and cloned voices. |
Output
The node outputs JSON data containing the parsed results from the image(s). The structure typically includes recognized text, metadata, or other analysis results depending on the model and input. If the "simplify" option is enabled, the output will be a streamlined version focusing on essential information.
If binary data is involved, it is expected to be handled via the specified binary property name, but the node primarily returns JSON-formatted analysis results rather than raw binary output.
Dependencies
- Requires access to the 阿里通义 (Ali Tongyi) API or service endpoint.
- Needs appropriate API authentication credentials configured in n8n (e.g., an API key or token).
- For voice-related features, uses a list of predefined voices accessible via a search method integrated into the node.
Troubleshooting
- Invalid Image URL: Ensure that the URLs provided are valid and accessible publicly. Invalid or inaccessible URLs will cause errors.
- Binary Data Issues: When using base64 input, verify that the binary data is correctly encoded and the binary property name matches the input field.
- Authentication Errors: Confirm that the API credentials are correctly set up and have sufficient permissions.
- Unsupported Model or Agent: If the model string is incorrect or unsupported, the node may fail to process the request.
- Simplify Option Effects: If output seems incomplete, try toggling the "simplify" option to get full detailed responses.
Links and References
- 阿里通义官方文档 (Ali Tongyi official documentation - Chinese)
- n8n Documentation on Custom Nodes: https://docs.n8n.io/integrations/creating-nodes/
- General image parsing and OCR concepts: https://en.wikipedia.org/wiki/Optical_character_recognition