The USTC-NEL Speech Translation system at IWSLT 2018
This work addresses speech translation for evaluation benchmarks, but it is incremental as it uses conventional pipeline methods.
The paper tackled the speech translation task at IWSLT 2018 by developing a pipeline system with speech recognition, post-processing, and machine translation modules, achieving a 14.9 BLEU improvement over the baseline system from KIT.
This paper describes the USTC-NEL system to the speech translation task of the IWSLT Evaluation 2018. The system is a conventional pipeline system which contains 3 modules: speech recognition, post-processing and machine translation. We train a group of hybrid-HMM models for our speech recognition, and for machine translation we train transformer based neural machine translation models with speech recognition output style text as input. Experiments conducted on the IWSLT 2018 task indicate that, compared to baseline system from KIT, our system achieved 14.9 BLEU improvement.