Receipt Scan API

The Lucidtech Receipt Scan API extracts key information from scanned, photographed or digital receipts with market leading precision.

Lucidtech - Invoice Scan API

Exploit your own data

Many enterprises have over the years built up thousands of receipts in their systems that can efficiently be exploited by our machine learning technology. By training our machine learning models on your business' data, we can customize the API to your needs.

Results you can trust

LucidReceipt API extracts a field-wise confidence estimation that can be used to accurately predict the likelihood that the extracted information is correct. This makes sure that the extracted information is actionable.

Simple integration

Our API can be integrated in existing software in a matter of hours. We provide SDKs for all the major languages upon request.

Why choose Lucidtech

Accurate

Huge amounts of data fuel our AI, so that we can provide accurate predictions.

Fast

Don't gamble with your users' patience. Our API responds within seconds.

Secure

We work hard to implement the latest security best practices to handle your data responsibly.

Frequently Asked

Here you can find the answers for the most frequently asked questions by our customers. If you have other questions, please do not hesitate to contact us at hello@lucidtech.ai.

1. What is a custom data training and how does it work?

Custom data training means that we train our machine learning models on customer data. This enables us to tailor our data extraction model to the customer's needs. After the training process the trained model is exposed as a customer-specific REST API that the customer can integrate with.

In order to do custom training, we need a reasonably large data set. The exact number depends on the variation of the data, but even in relatively simple cases we typically need at least a few thousand labeled invoices. The training process usually takes 1-4 weeks.

2. What languages are supported?

The LucidReceipt API can be trained on any language. We have a pre-trained model that is trained on a variety of European and American receipts. Other languages and/or receipt formats are supported through custom data training.

3. What kind of fields does the API extract?

Our pre-trained Receipt API only extracts date and total amount. If you would like to extract other fields, we can create a custom API for your need.

4. What is the accuracy?

Accuracy largely depends on how similar the test data and the training data is. We encourage you to run tests on your own data in order to get an accurate estimate of the accuracy.

5. What is the pricing?

The pricing depends on the volume, document type and your need for customization. Please contact us with a description of your case for a price offer.

Get started today

Try our demo, and see why developers choose us.