%Aigaion2 BibTeX export from Idiap Publications %Saturday 21 December 2024 05:35:44 PM @INPROCEEDINGS{Juan_EMNLP_2023, author = {Zuluaga-Gomez, Juan and Huang, Zhaocheng and Niu, Xing and Srinavasan, Sundararajan and Mathur, Prashant and Thompson, Brian and Federico, Marcello}, projects = {Idiap}, month = dec, title = {End-to-End Single-Channel Speaker-Turn Aware Conversational Speech Translation}, booktitle = {The 2023 Conference on Empirical Methods in Natural Language Processing (EMNLP)}, series = {1}, volume = {1}, number = {1}, year = {2023}, location = {Singapore}, url = {https://arxiv.org/abs/2311.00697}, abstract = {Conventional speech-to-text translation (ST) systems are trained on single-speaker utterances, and they may not generalize to real-life scenarios where the audio contains conversations by multiple speakers. In this paper, we tackle single-channel multi-speaker conversational ST with an end-to-end and multi-task training model, named Speaker-Turn Aware Conversational Speech Translation, that combines automatic speech recognition, speech translation and speaker turn detection using special tokens in a serialized labeling format. We run experiments on the Fisher-CALLHOME corpus, which we adapted by merging the two single-speaker channels into one multi-speaker channel, thus representing the more realistic and challenging scenario with multi-speaker turns and cross-talk. Experimental results across single- and multi-speaker conditions and against conventional ST systems, show that our model outperforms the reference systems on the multi-speaker condition, while attaining comparable performance on the single-speaker condition. We release scripts for data processing and model training.}, pdf = {https://publications.idiap.ch/attachments/papers/2023/Juan_EMNLP_2023.pdf} }