Segmentation of the incoming speech stream and translating segments incrementally is a commonly used technique that improves latency in spoken language translation. Previous work of Oda et al. 2014  has explored creating training data for segmentation by finding segments that maximize translation quality with a user-defined bound on segment length.In this work, we provide a new algorithm that uses Pareto-optimality to find good segment boundaries that can balance the trade-off between latency versus translation quality. We compare against the state-of-the-art greedy algorithm from Oda et al. 2014. Our experimental results show that we can improve latency by up to 12% without harming theBleuscore for the same average segment length. Another benefit is that for any segment size,Pareto-optimal segments maximize both latency and translation quality.
Copyright is held by the author.
This thesis may be printed or downloaded for non-commercial research and scholarly purposes.
Supervisor or Senior Supervisor
Thesis advisor: Sarkar, Anoop
Member of collection