Lingvo: a Modular and Scalable Framework for Sequence-to-Sequence Modeling

TL;DR

Lingvo is a flexible, modular TensorFlow framework designed for sequence-to-sequence modeling, supporting distributed training, quantized inference, and extensive research utilities, facilitating collaborative deep learning research.

Contribution

It introduces a highly modular, scalable framework for sequence-to-sequence models with centralized configuration, extensibility, and built-in support for advanced training features.

Findings

Used in over 20 research papers

Supports distributed training and quantized inference

Provides extensive utilities and modular components

Abstract

Lingvo is a Tensorflow framework offering a complete solution for collaborative deep learning research, with a particular focus towards sequence-to-sequence models. Lingvo models are composed of modular building blocks that are flexible and easily extensible, and experiment configurations are centralized and highly customizable. Distributed training and quantized inference are supported directly within the framework, and it contains existing implementations of a large number of utilities, helper functions, and the newest research ideas. Lingvo has been used in collaboration by dozens of researchers in more than 20 papers over the last two years. This document outlines the underlying design of Lingvo and serves as an introduction to the various pieces of the framework, while also offering examples of advanced features that showcase the capabilities of the framework.