The Interpretor
A powerful NLP API

Use the artificial intelligence of the Interpretor, our NLP engine, to find out what's in customer feedback, news articles, social media posts, blogs, emails, or legal documents in your archives. Discover how powerful, customizable, and scalable it is.

See our demo now!
Interpretor

Trusted by

ŠKODA
Keboola
Vocalls
Globus
amio
Newton Media

Customizable

Use your own dictionaries, labels, and categories. Have your own private analyzer tuned to your data and your needs.

Clean and easy-to-use API

Use our REST API to integrate the Interpretor capabilities into your workflow, or use an SDK.

Out-of-the-box domains

Preconfigured for news, hospitality, banking and more.

Learn more

Keboola integration

Calling the Interpretor from Keboola Connection is a matter of a few clicks.

Learn more

Features

Preprocessing & Low-Level NLP Features

Format extraction

Extract text from various formats, including HTML, PDF, and MS Word.

Correction

Perform basic spelling correction on all the words in the document.

Diacritics correction

Perform diacritization on all the words in the document

Language

Identify which of the supported 30+ languages the document was written in.

A great pizza English
Výborná pizza Czech

Morphology

Find the appropriate dictionary form and part of speech for each word.

Morphology

Relations

Summarize what actions are expressed in the text (i.e., who did what to whom)

Relations

High-Level NLP Features

Named and general entities

Find people, locations, dates, times, phone numbers, colors, economic terms, menu items, etc. Link them to Wikipedia, Open Streets Map, or your own database.

President George Washington visited Paris on July 13

Sentiment

Find out what the attitude of the text’s author was. We give you the sentiment of the whole review, as well as that of individual sentences and entities.

On Saturday, the pizza was great, but
the waiter was so annoying.

Attributes

Sentiment gives you a numerical value. With attributes, you get a better understanding of what words your customers use to describe your employees, your products, etc.

Pizza great soggy pricey
Assistant nice helpful knowledgeable

Intent

Identify a customer’s expected problem or desired action and connect your chatbot.

Lost credit card Buy plane ticket

Categories

Sort texts into predefined categories. Use industry-standard categories such as IAB, IPTC, our out-of-the-box categories for restaurant VoC, or your own custom labels.

Technical support Billing Cancelations

Tags

Categories work for texts talking about what you anticipate. To find unexpected information, use our tags or keywords.

Cancellation Safe car Terrible cook

Supported languages

The public API supports English, Czech, Dutch, French, German, and Slovak.

Support for Polish and Spanish is in beta.

We are able to add support for a new language in two weeks if required.

How it works

Technicalities

The Interpretor uses a pipeline of modules to analyze each document. We use a combination of machine learning modules (neural networks, support vector machines, conditional random fields, maximum entropy, etc.), rule-based models, and large knowledge bases to obtain the best possible results. Both the pipeline and the individual modules can be customized for a particular use-case.

Step 1

Preprocessing

  • Format conversion
  • Language detection
  • Sentence segmentation
  • Tokenization

Step 2

Grammar

  • Correction
  • Lemmatization
  • Morphological tagging
  • Syntactic structure

Step 3

Entities

  • Discovery
  • Standardization
  • Disambiguation
  • Linking

Step 4

Other

  • Relations
  • Sentiment
  • Categorization
  • Intent
What else?

Resources

Everything you need to integrate Geneea’s Interpretor into your solution

Quick Start Full Documentation Download SDKs Interactive Demo Interactive API Reference
GET FREE API TRIAL

The Latest from Geneea

Newsrooms
Newsrooms Can Save Thousands of Hours with a RAG-Based AI Research System

For journalists, the most time-consuming part of writing an article is finding information. Gathering context, looking up numbers, and verifying facts can take hours—pulling them away from in-depth analysis, interviews, and uncovering new stories. As a result, newsrooms may spend hundreds—if not thousands—of hours every month on these tasks. With the rise of AI, many...
Geneea’s
Geneea’s AI Spotlight #11

The eleventh edition of our newsletter on Large Language Models is here. In this edition, we explore • Stanford's AI Index Report, • a mix of releases, • AI in journalism, and • tips for improving model answers.
Geneea’s
Geneea’s AI Spotlight #10

The tenth edition of our newsletter on Large Language Models is here. In this edition, we explore • how generative AI is used by companies and people, • new models and infrastructure, • experience from real deployment, • challenges, misuse, and • ColBERT, an efficient approach to search.
See all Geneea news