Catalogue of Natural Language Generation Systems

Welcome to the Systems section of NLG-Wiki.org — your dedicated reference point for exploring the wide range of tools, frameworks and platforms used in natural language generation (NLG). Whether you’re a researcher, developer or student, this section gives you an organised overview of NLG systems sorted by architecture, application domain and licence type.

What you’ll find here

• Open-source toolkits such as template-based generators, hybrid systems and neural sequence-to-sequence models.

• Commercial platforms used for automated reporting, summarisation, dialogue systems and narrative generation.

• Research systems built in academia, with links to papers, datasets and code.

• Historical engines that trace the evolution of NLG from rule-based to modern neural approaches.

How the catalogue is structured

Each system entry includes:

• System name and developers.

• Primary architecture (e.g., rule-based, statistical, neural).

• Input / output types (e.g., data-to-text, image-captioning, dialogue).

• Licence and availability (open-source, academic, commercial).

• Key publication references and links to source code or demos (when available).

Who this section is for

• Researchers looking for baseline systems, comparative studies or implementation details.

• Developers and engineers selecting NLG engines for projects in business intelligence, media, healthcare and beyond.

• Students and newcomers wanting to gain an overview of the NLG landscape and discover working systems.

Why catalogue NLG systems?

The field of natural language generation encompasses a vast variety of methods — from hand-crafted templates to large pretrained language models. A structured catalogue helps you navigate this diversity:

• Compare systems by architecture, domain and maturity.

• Find reproducible implementations and code for experimentation.

• Understand how different design choices (e.g., planning, lexical choice, realisation) are realised in practice.

How to use this page effectively

1. Use the filters or categories to locate systems by architecture or task.

2. Examine the system details for input/output type and code links.

3. Follow publication references to dive deeper into methodology and evaluation.

4. Contribute updates or new system entries — the wiki is continuously evolving with community input.