NeMo: a toolkit for building AI applications using Neural Modules

TL;DR

NeMo is an open-source Python toolkit that simplifies building AI applications by enabling modular, reusable neural components with semantic correctness, supporting speech and language tasks with distributed training capabilities.

Contribution

It introduces a neural module-based framework with a type system for semantic correctness, facilitating flexible, reusable AI component assembly for speech and language applications.

Findings

Supports distributed training and mixed precision on NVIDIA GPUs

Provides extendable collections for speech recognition and NLP

Enhances modularity and reusability in AI application development

Abstract

NeMo (Neural Modules) is a Python framework-agnostic toolkit for creating AI applications through re-usability, abstraction, and composition. NeMo is built around neural modules, conceptual blocks of neural networks that take typed inputs and produce typed outputs. Such modules typically represent data layers, encoders, decoders, language models, loss functions, or methods of combining activations. NeMo makes it easy to combine and re-use these building blocks while providing a level of semantic correctness checking via its neural type system. The toolkit comes with extendable collections of pre-built modules for automatic speech recognition and natural language processing. Furthermore, NeMo provides built-in support for distributed training and mixed precision on latest NVIDIA GPUs. NeMo is open-source https://github.com/NVIDIA/NeMo