We describe a novel two-way speech-to-speech (S2S) translation system that actively detects a wide variety of common error types and resolves them through user-friendly dialog with the user(s). We present algorithms for detecting out-of-vocabulary (OOV) named entities and terms, sense ambiguities, homophones, idioms, ill-formed input, etc. and discuss novel, interactive strategies for recovering from such errors. We also describe our approach for prioritizing different error types and an extensible architecture for implementing these decisions. We demonstrate the efficacy of our system by presenting analysis on live interactions in the English-to-Iraqi Arabic direction that are designed to invoke different error types for spoken language translation. Our analysis shows that the system can successfully resolve 47% of the errors, resulting in a dramatic improvement in the transfer of problematic concepts.
Cite as: Prasad, R., Kumar, R., Ananthakrishnan, S., Chen, W., Hewavitharana, S., Roy, M., Choi, F., Challenner, A., Kan, E., Neelakantan, A., Natarajan, P. (2012) Active error detection and resolution for speech-to-speech translation. Proc. International Workshop on Spoken Language Translation (IWSLT 2012), 150-157
@inproceedings{prasad12_iwslt,
author={Rohit Prasad and Rohit Kumar and Sankaranarayanan Ananthakrishnan and Wei Chen and Sanjika Hewavitharana and Matthew Roy and Frederick Choi and Aaron Challenner and Enoch Kan and Arvind Neelakantan and Prem Natarajan},
title={{Active error detection and resolution for speech-to-speech translation}},
year=2012,
booktitle={Proc. International Workshop on Spoken Language Translation (IWSLT 2012)},
pages={150--157}
}