Espresso is an open-source, modular, extensible end-to-end neural automatic speech recognition (ASR) toolkit based on the deep learning library PyTorch and the popular neural machine translation toolkit fairseq. Espresso supports distributed training across GPUs and computing nodes, and features various decoding approaches commonly employed in ASR, including look-ahead word-based language model fusion, for which a fast, parallelized decoder is implemented.
We provide state-of-the-art training recipes for the following speech datasets:
- September 2019: We are in an effort of isolating Espresso from fairseq, resulting in a standalone package that can be directly
pip installed.
- PyTorch version >= 1.1.0
- Python version >= 3.5
- For training new models, you'll also need an NVIDIA GPU and NCCL
- For faster training install NVIDIA's apex library with the
--cuda_extoption
Currently Espresso only support installing from source.
To install fairseq from source and develop locally:
git clone https://github.com/freewym/espresso
cd espresso
pip install --editable .
pip install kaldi_io
pip install sentencepiece
cd speech_tools; make KALDI=<path/to/a/compiled/kaldi/directory>add your Python path to PATH variable in examples/asr_<dataset>/path.sh, the current default is ~/anaconda3/bin.
kaldi_io is required for reading kaldi scp files. sentencepiece is required for subword pieces training/encoding. Kaldi is required for data preparation, feature extraction and scoring for some datasets (e.g., Switchboard).
Espresso is MIT-licensed.
Please cite Espresso as:
@inproceedings{wang2019espresso,
title = {Espresso: A Fast End-to-end Neural Speech Recognition Toolkit},
author = {Yiming Wang and Tongfei Chen and Hainan Xu
and Shuoyang Ding and Hang Lv and Yiwen Shao
and Nanyun Peng and Lei Xie and Shinji Watanabe
and Sanjeev Khudanpur},
booktitle = {2019 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU)},
year = {2019},
}