AI处理收据 Make

PDF4me AI处理收据是一个 Make 读取收据的模块 PDF 或者，它可以拍摄照片并返回结构化的费用数据，涵盖商家详情、明细项目、小计、税款和总计。使用它可以自动提交费用、跟踪报销或批量处理税款扣除，无需手动将收据数据重新输入电子表格。

本模块的功能

PDF4me人工智能处理收据 阅读收据 PDF 或图像在 Make 该工具可处理各种场景，并返回结构化的费用数据包，包括商家名称、地址、电话和网站，以及每项费用明细（数量、价格、小计、税额、总计和支付方式）。它支持 11 种收据类别（餐饮、酒店、燃油、医疗保健、订阅服务等），无需模板配置即可处理电子收据和纸质收据照片。

相关博客文章

目前尚无关于此功能的博客文章——敬请期待。

在此期间，您可以浏览 PDF4me 博客，查看适用于各平台的教程和工作流程。

访问博客 →

验证您的身份 API 要求

每一个 PDF4me 模块 Make 需要有效 联系创建或选择一个可以容纳您的 PDF4me API 关键在于确保该方案能够安全地验证收据提取请求。

Get your API Key

您不容错过的重要事实

11 收据类别

将收据类型设置为餐饮、酒店、用品、燃油、交通、医疗保健、订阅等，或留空以进行通用 AI 检测。设置类型可以提高该特定收据类别的字段识别准确率。

完整的商户和商品明细数据

返回商家名称、地址、电话、网站、商店编号、每项商品的数量和价格、小计、税金、总计和付款方式，所有这些都包含在模块输出包中，可以映射到下游。

独立提取开关

分析商品、提取商家信息和计算总计是独立的开关，您可以禁用任何不需要的部分，以保持输出包简洁并加快处理速度。

PDF4me AI处理收据模块提取结构化的费用数据 Make 设想。

参数

必需的： 连接方式、收据名称、文档和输入数据类型。其他字段均为可选，请仅启用工作流程所需的字段。

范围	必需的	它的作用	例子
Connection	Required	PDF4me API connection. Click Add and paste your API key if connecting for the first time.	Your PDF4me connection
Receipt Name	Required	Filename of the receipt including extension: used for format detection and processing accuracy.	receipt_march.pdf
Document	Required	Binary receipt file mapped from a preceding module such as Dropbox Get File or Gmail Get Attachment.	[Buffer from Get File]
Input Data Type	Required	Format of the input: Binary Data, Base64 String, or URL: choose based on how your preceding module delivers the file.	Binary Data
Receipt Type	Optional	Category hint for the AI: meal, supplies, hotel, fuel and energy, transportation, communication, subscriptions, entertainment, training, healthcare, or generic.	meal
Analyze Items	Optional	Set to Yes to extract individual line items with name, quantity, unitPrice, totalPrice, category, and description.	Yes
Extract Merchant Info	Optional	Set to Yes to extract merchant name, address, phone, website, and store number.	Yes
Calculate Totals	Optional	Set to Yes to extract subtotal, tax amount, and total amount from the receipt.	Yes
Custom Field Keys	Optional	Optional array of extra field names to extract beyond standard receipt data.	["loyalty_points"]

输出

场地	类型	它包含什么
merchantName	String	Name of the store or merchant printed on the receipt.
merchantAddress	String	Merchant's address, when Extract Merchant Info is enabled and the receipt shows one.
subtotal	Number	Pre-tax subtotal amount, populated when Calculate Totals is enabled.
taxAmount	Number	Tax amount extracted from the receipt, populated when Calculate Totals is enabled.
totalAmount	Number	Final total amount charged, populated when Calculate Totals is enabled.
paymentMethod	String	Payment method shown on the receipt, such as cash or card, when detected.
analyzeItems	Array	Line items, each with name, quantity, unitPrice, totalPrice, category, and description, populated when Analyze Items is enabled.
merchantInfoList	Array	Additional merchant detail fields, populated when Extract Merchant Info is enabled.

如何处理收据 Make 和 PDF4me WHO？

1. 添加 PDF4me → AI处理收据 致你 Make 设想。
2. 选择 联系 （或点击） 添加 用你的 PDF4me API 钥匙）。
3. 放 输入数据类型 到 Binary Data （大多数云存储源返回二进制文件）。
4. 将上一个模块中的收据二进制文件映射并设置 收据名称 包括文件扩展名。
5. （可选） 收据类型 （例如。 meal 或者 hotel）并启用 分析项目， 提取商家信息， 和 计算总数 根据需要。
6. 运行场景：提取的费用数据将出现在输出包中，准备进行下游映射。

何时应该使用AI处理收据？

典型配置Common Make scenario patterns using AI-Process Receipt.

员工费用申报

1. Gmail Watch 检测到员工发送了一封带有收据附件的新邮件。
2. 获取附件功能检索二进制文件，然后 AI 处理收据功能提取商户、金额和明细项目。
3. QuickBooks 创建费用功能会将收据作为已分类的费用发布，所有字段均已预先填写。
4. Slack 通知财务团队，已记录一笔新的费用待审核。

酒店住宿报销

1. Dropbox 监视文件夹检测到出差员工上传的酒店收据。
2. AI 处理收据（收据类型：酒店）提取每晚房价、日期、税费和总计。
3. Google Sheets 添加行会将所有字段记录到报销跟踪表中。
4. Slack 会向审批人发送通知，其中包含总金额和提交日期。

年度税收扣除批次

1. 定时触发器会列出 Google 云端硬盘文件夹中该纳税年度的所有收据。
2. 迭代器遍历每个文件，获取文件 → AI 处理收据。
3. 汇总器按收据类型（餐饮、旅行、用品）汇总总额。
4. Google Sheets 可以导出用于报税的类别总计，所有操作只需一次即可完成。

实用技巧

The three extraction toggles are independent

Analyze Items, Extract Merchant Info, and Calculate Totals can be turned on or off separately. Disable any section you do not need to keep the output bundle smaller and processing faster.

Receipt Type is a hint, not a requirement

Leaving Receipt Type blank still works, the AI falls back to generic detection. Set it only when you know the category and want higher field-recognition accuracy.

Input Data Type must match what the prior module sends

Binary Data, Base64 String, and URL are not interchangeable. Check what your Dropbox, Gmail, or Google Drive module actually outputs before setting this field.

Custom Field Keys extends extraction beyond the defaults

Use this array to pull additional named fields, such as loyalty program numbers, that are not part of the standard merchant or line-item output.

Line items live in a nested array

analyzeItems is an array of objects, not flat fields. Use an Iterator module downstream to process each line item individually, for example when posting to an accounting system.

速查表

提取的费用字段直接导入到记录系统中。有关收据和证明文件的保留期限，请参阅[此处应插入相关内容]。 美国国税局记录保存指南 面向美国企业。有关将纸质文档数字化以确保其在……之前仍可读的一般指南 OCR 以及人工智能提取工具，请参见 联邦机构数字化指南倡议。

场地	价值
Input Data Type	Binary Data (common), Base64 String, or URL
Receipt Type	meal, hotel, fuel and energy, or leave blank
Analyze Items	Yes, to get the analyzeItems line-item array
Extract Merchant Info	Yes, to get merchantName, merchantAddress, merchantInfoList
Calculate Totals	Yes, to get subtotal, taxAmount, totalAmount

常见问题解答

What receipt types does AI-Process Receipt support?+

The module supports 11 receipt categories: meal, supplies, hotel, fuel and energy, transportation, communication, subscriptions, entertainment, training, healthcare, and generic. Setting the Receipt Type improves AI extraction accuracy for that category. Leave it empty for generic detection on any receipt.

What fields are returned in the analyzeItems output?+

Each item in the analyzeItems array contains: name, quantity, unitPrice, totalPrice, category, and description: covering every line item on the receipt. Enable Analyze Items in the module settings to populate this array. Use an Iterator module to loop through individual line items downstream.

Do I need to set Receipt Type for the module to work?+

No: Receipt Type is optional. The AI processes any receipt without it using generic detection. Setting it (for example hotel or meal) improves field recognition accuracy for that specific receipt category without any other configuration changes needed.

Can AI-Process Receipt read photographed receipts?+

Yes. PNG, JPG, and JPEG inputs are fully supported: including smartphone photos of paper receipts. For best results: clear, even lighting and the full receipt in frame. Crumpled or partially obscured receipts may reduce extraction accuracy.

Where do merchant and financial details appear in Make?+

All fields appear in the module output bundle. Click any downstream module field and map keys directly: for example merchantName, totalAmount, or analyzeItems[].unitPrice. Use an Iterator module to process individual line items from the analyzeItems array one by one.

How long should I keep digitized receipt records for taxes?+

Retention periods vary by jurisdiction and expense type. Extracted fields from this module, merchantName, totalAmount, date, and category, can be logged to a spreadsheet or database automatically to build that record without manual entry.

相关操作

AI发票解析器

利用人工智能从发票中提取供应商详细信息、明细项目和总计，完善收据处理，实现全面的费用管理。

AI处理工资单

从工资单中提取工资数据，包括总工资、税款、年初至今总计等，并将其打包成结构化的数据包。

分类文件

在将文档路由到正确的AI提取模块之前，自动识别文档类型。

在其他平台上执行相同任务

n8n

与人工智能收据提取相同的 n8n 在您的自托管或云工作流程中，没有节点。 Zapier， Power Automate或独立 API 该模块的页面尚未存在。

获取帮助

Make IntegrationGet your API KeyMake DocumentationModule Reference