In this paper, we present TED-LIUM release 3 corpus dedicated to speech recognition in English, that multiplies by more than two the available data to train acoustic models in comparison with TED-LIUM 2. We present the recent development on Automatic Speech Recognition (ASR) systems in comparison with the two previous releases of the TED-LIUM Corpus from 2012 and 2014. We demonstrate that, passing from 207 to 452 hours of transcribed speech training data is really more useful for end-to-end ASR systems than for HMM-based state-of-the-art ones, even if the HMM-based ASR system still outperforms end-to-end ASR system when the size of audio training data is 452 hours, with respectively a Word Error Rate (WER) of 6.6% and 13.7%. Last, we propose two repartitions of the TED-LIUM release 3 corpus: the legacy one that is the same as the one existing in release 2, and a new one, calibrated and designed to make experiments on speaker adaptation. Like the two first releases, TED-LIUM 3 corpus will be freely available for the research community.
@article{arxiv.1805.04699,
title = {TED-LIUM 3: twice as much data and corpus repartition for experiments on speaker adaptation},
author = {François Hernandez and Vincent Nguyen and Sahar Ghannay and Natalia Tomashenko and Yannick Estève},
journal= {arXiv preprint arXiv:1805.04699},
year = {2019}
}
Comments
Submitted to SPECOM 2018, 20th International Conference on Speech and Computer; TED-LIUM 3 corpus available on https://lium.univ-lemans.fr/en/ted-lium3/