OpenSeq2Seq – TensorFlow-based toolkit for sequence-to-sequence models

OpenSeq2Seq is a toolkit for distributed and mixed precision training of sequence-to-sequence models:

• Machine translation (GNMT, Transformer, ConvS2S, …). These models have been trained with BPE vocabulary used for text tokenization.
• Speech recognition (DeepSpeech2, Wave2Letter, Jasper, …).
• Speech commands (RN-50, Jasper). Automatic speech recognition (ASR) systems can be built using a number of approaches depending on input data type, intermediate representation, model’s type and output post-processing. OpenSeq2Seq is currently focused on end-to-end CTC-based models (like original DeepSpeech model).
• Speech synthesis (Tacotron2, Tacotron2 GST, WaveNet, Centaur, …).
• Language model (LSTM with WikiText-2, LSTM with WikiText-103).
• Sentiment analysis (SST, IMDB, …).
• Image classification, a mixed precison replica of TensorFlow ResNet-50.

OpenSeq2Seq main goal is to allow researchers to most effectively explore various sequence-to-sequence models. The efficiency is achieved by fully supporting distributed and mixed-precision training.

OpenSeq2Seq is built using TensorFlow and provides all the necessary building blocks for training encoder-decoder models for neural machine translation, automatic speech recognition, speech synthesis, and language modeling.

Speech-to-text workflow uses some parts of Mozilla DeepSpeech project.

This is a research project, not an official NVIDIA product.

Features include:

• Models for:
  1. Neural Machine Translation.
  2. Automatic Speech Recognition.
  3. Speech Synthesis.
  4. Language Modeling.
  5. NLP tasks (sentiment analysis).
• Data-parallel distributed training:
  1. Multi-GPU.
  2. Multi-node.
• Mixed precision training for NVIDIA Volta/Turing GPUs.
• Supports two modes for parallel training: simple multi-tower approach and Horovod-based approach.
• Supports two new optimizers: Layer-wise Adaptive Rate Control (LARC) and NovoGrad. NovoGrad is a first-order SGD-based algorithm, which computes second moments per layer instead of per weight as in Adam.
• Mixed precision with existing models.
• Interactive infer – a mode that makes it easy to demo trained models.

Website: nvidia.github.io/OpenSeq2Seq
Support: GitHub Code Repository
Developer: NVIDIA
License: Apache License 2.0

OpenSeq2Seq is written in Python. Learn Python with our recommended free books and free tutorials.

Return to Speech Recognition Tools

Popular series
	The largest compilation of the best free and open source software in the universe. Each article is supplied with a legendary ratings chart helping you to make informed decisions.
	Hundreds of in-depth reviews offering our unbiased and expert opinion on software. We offer helpful and impartial information.
	The Big List of Active Linux Distros is a large compilation of actively developed Linux distributions.
	Replace proprietary software with open source alternatives: Google, Microsoft, Apple, Adobe, IBM, Autodesk, Oracle, Atlassian, Corel, Cisco, Intuit, and SAS.
	Awesome Free Linux Games Tools showcases a series of tools that making gaming on Linux a more pleasurable experience. This is a new series.
	Machine Learning explores practical applications of machine learning and deep learning from a Linux perspective. We've written reviews of more than 40 self-hosted apps. All are free and open source.
	New to Linux? Read our Linux for Starters series. We start right at the basics and teach you everything you need to know to get started with Linux.
	Alternatives to popular CLI tools showcases essential tools that are modern replacements for core Linux utilities.
	Essential Linux system tools focuses on small, indispensable utilities, useful for system administrators as well as regular users.
	Linux utilities to maximise your productivity. Small, indispensable tools, useful for anyone running a Linux machine.
	Surveys popular streaming services from a Linux perspective: Amazon Music Unlimited, Myuzi, Spotify, Deezer, Tidal.
	Saving Money with Linux looks at how you can reduce your energy bills running Linux.
	Home computers became commonplace in the 1980s. Emulate home computers including the Commodore 64, Amiga, Atari ST, ZX81, Amstrad CPC, and ZX Spectrum.
	Now and Then examines how promising open source software fared over the years. It can be a bumpy ride.
	Linux at Home looks at a range of home activities where Linux can play its part, making the most of our time at home, keeping active and engaged.
	Linux Candy reveals the lighter side of Linux. Have some fun and escape from the daily drudgery.
	Getting Started with Docker helps you master Docker, a set of platform as a service products that delivers software in packages called containers.
	Best Free Android Apps. We showcase free Android apps that are definitely worth downloading. There's a strict eligibility criteria for inclusion in this series.
	These best free books accelerate your learning of every programming language. Learn a new language today!
	These free tutorials offer the perfect tonic to our free programming books series.
	Linux Around The World showcases usergroups that are relevant to Linux enthusiasts. Great ways to meet up with fellow enthusiasts.
	Stars and Stripes is an occasional series looking at the impact of Linux in the USA.