Presenting Simultaneous Translation in Limited Space
This addresses a deployment challenge for users of automatic simultaneous translation systems, but it is incremental as it focuses on presentation rather than core translation improvements.
The paper tackles the problem of presenting simultaneous translation subtitles in limited space, such as on a TV screen, by providing an algorithm for subtitling and proposing a method to estimate usability based on quality, latency, and stability, including an improved latency measure.
Some methods of automatic simultaneous translation of a long-form speech allow revisions of outputs, trading accuracy for low latency. Deploying these systems for users faces the problem of presenting subtitles in a limited space, such as two lines on a television screen. The subtitles must be shown promptly, incrementally, and with adequate time for reading. We provide an algorithm for subtitling. Furthermore, we propose a way how to estimate the overall usability of the combination of automatic translation and subtitling by measuring the quality, latency, and stability on a test set, and propose an improved measure for translation latency.