NeurST: Neural Speech Translation Toolkit

TL;DR

NeurST is an open-source toolkit designed to facilitate end-to-end neural speech translation research by providing easy-to-use features, benchmarks, and experimental results for various datasets.

Contribution

It introduces a comprehensive framework for neural speech translation, including recipes for data processing, training, and evaluation, with reliable benchmarks for future research.

Findings

Provides step-by-step recipes for speech translation tasks

Establishes reliable benchmarks for various datasets

Demonstrates competitive experimental results

Abstract

NeurST is an open-source toolkit for neural speech translation. The toolkit mainly focuses on end-to-end speech translation, which is easy to use, modify, and extend to advanced speech translation research and products. NeurST aims at facilitating the speech translation research for NLP researchers and building reliable benchmarks for this field. It provides step-by-step recipes for feature extraction, data preprocessing, distributed training, and evaluation. In this paper, we will introduce the framework design of NeurST and show experimental results for different benchmark datasets, which can be regarded as reliable baselines for future research. The toolkit is publicly available at https://github.com/bytedance/neurst/ and we will continuously update the performance of NeurST with other counterparts and studies at https://st-benchmark.github.io/.