Skip to content

Natural Language Generation

Generation of text used in data-to-text, conversational agents, and narrative generation applications.

Source

74 resources imported from accelerated-text/awesome-nlg, dedicated to the public domain (CC0 / Unlicense). Thanks to its maintainers and contributors. Part of the KEEBOX Library · Category: Computer Science.

Datasets

  • Alex Context NLG Dataset - A dataset for NLG in dialogue systems in the public transport information domain.
  • Box-score data - This dataset consists of (human-written) NBA basketball game summaries aligned with their corresponding box- and line-scores.
  • E2E - This shared task focuses on recent end-to-end (E2E), data-driven NLG methods, which jointly learn sentence planning and surface realisation from non-aligned data.
  • Neural-Wikipedian - The repository contains the code along with the required corpora that were used in order to build a system that "learns" how to generate English biographies for Semantic Web triples.
  • WeatherGov - Computer-generated weather forecasts from weather.gov (US public forecast), along with corresponding weather data.
  • WebNLG - The enriched version of the WebNLG - a resource for evaluating common NLG tasks, including Discourse Ordering, Lexicalization and Referring Expression Generation.
  • WikiBio - wikipedia biography dataset - This dataset gathers 728,321 biographies from wikipedia. It aims at evaluating text generation algorithms.
  • The Schema-Guided Dialogue Dataset - The Schema-Guided Dialogue (SGD) dataset consists of over 20k annotated multi-domain, task-oriented conversations between a human and a virtual assistant.
  • The Wikipedia company corpus - Company descriptions collected from Wikipedia. The dataset contains semantic representations, short, and long descriptions for 51K companies in English.
  • YelpNLG - YelpNLG provides resources for natural language generation of restaurant reviews.

Dialog

  • Chatito - Generate datasets for AI chatbots, NLP tasks, named entity recognition or text classification models using a simple DSL!
  • NNDIAL - NNDial is an open source toolkit for building end-to-end trainable task-oriented dialogue models.
  • Plato - This is the Plato Research Dialogue System, a flexible platform for developing conversational AI agents.
  • RNNLG - RNNLG is an open source benchmark toolkit for Natural Language Generation (NLG) in spoken dialogue system application domains.
  • TGen - Statistical NLG for spoken dialogue systems.

Evaluation

Grammar

  • OpenCCG - OpenCCG library for parsing and realization with CCG.
  • GrammaticalFramework - A programming language for multilingual grammar applications.
  • EasyCCG - CCG: All combinators, common grammar format, parsing to logical form, parameter estimation for probabilistic CCG.
  • CCG Lab - All combinators, common grammar format, parsing to logical form, parameter estimation for probabilistic CCG.
  • CCGweb - A Web platform for parsing and annotation.

Libraries

  • Cron Expression Descriptor - A .NET library that converts cron expressions into human readable descriptions.
  • Number Words - Convert a number to an approximated text expression: from '0.23' to 'less than a quarter'.
  • Writebot - A NodeJS library that makes it easier to use GPT-3 by using presets.

Narrative Generation

  • Random Story Generator - Using Natural Language Generation (NLG) to create a random short story.
  • Tracery - A story-grammar generation library for JavaScript.

Neural Natural Language Generation

  • aitextgen - A robust Python tool for text-based AI training and generation using GPT-2.
  • graph-2-text - Graph to sequence implemented in Pytorch combining Graph convolutional networks and opennmt-py.
  • Image Caption Generator - A Neural Network based generative model for captioning images using Tensorflow.
  • lightnlg - A minimalistic codebase for finetuning and interacting with NLG models using PyTorch Lightning.
  • PaperRobot: Incremental Draft Generation of Scientific Ideas - We present a PaperRobot who performs as an automatic research assistant.
  • PPLM - Plug and Play Language Model implementation. Allows to steer topic and attributes of GPT-2 models.
  • Question Generation using hugstransformers - Question generation is the task of automatically generating questions from a text paragraph.
  • Texar - Texar is a toolkit aiming to support a broad set of machine learning, especially natural language processing and text generation tasks.
  • textgenrnn - Easily train your own text-generating neural network of any size and complexity on any text dataset with a few lines of code.
  • This Word Does Not Exist - This is a project allows people to train a variant of GPT-2 that makes up words, definitions and examples from scratch.
  • Transformers - State-of-the-art Natural Language Processing for TensorFlow 2.0 and PyTorch.
  • Summary Generation From Structured Data - For converting information present in the form of structured data into natural language text.

Papers and Articles

Products

  • Accelerated Text - Automatically generate multiple natural language descriptions of your data varying in wording and structure.
  • RosaeNLG - An open-source library for node.js or client side (browser) execution, based on the Pug template engine, to generate texts in English, French, German and Italian.
  • Twine - An open-source tool for telling interactive, nonlinear stories.

Realizers

  • Genl - Surface realiser (part of a Natural Language Generation system) using Tree Adjoining Grammar.
  • JSrealB - A JavaScript bilingual text realizer for web development.
  • SimpleNLG - Java API for Natural Language Generation.
  • SimpleNLG DE - German version of SimpleNLG 4.
  • SimpleNLG-EnFr - SimpleNLG-EnFr 1.1 is a bilingual English/French adaption of SimpleNLG v4.2.

Templating Languages

  • calyx - A Ruby library for generating text with recursive template grammars.
  • nalgene - Natural language generation language.
  • StringTemplate - Java template engine (with ports for C##, Objective-C, JavaScript, Scala) for generating source code, web pages, emails, or any other formatted text output.

Videos