Joint Speech Translation and Named Entity Recognition
This addresses the need for more accurate and efficient contextual support in automatic translation systems, though it is incremental as it builds on existing direct speech translation models.
The paper tackled the problem of enriching speech translation output with named entity information by proposing multitask models that jointly perform speech translation and named entity recognition, which outperformed a cascade baseline by 0.4-1.0 F1 on NER without degrading translation quality or computational efficiency.
Modern automatic translation systems aim at place the human at the center by providing contextual support and knowledge. In this context, a critical task is enriching the output with information regarding the mentioned entities, which is currently achieved processing the generated translation with named entity recognition (NER) and entity linking systems. In light of the recent promising results shown by direct speech translation (ST) models and the known weaknesses of cascades (error propagation and additional latency), in this paper we propose multitask models that jointly perform ST and NER, and compare them with a cascade baseline. The experimental results show that our models significantly outperform the cascade on the NER task (by 0.4-1.0 F1), without degradation in terms of translation quality, and with the same computational efficiency of a plain direct ST model.