Attention as a guide for Simultaneous Speech Translation

Autor:	Sara Papi, Matteo Negri, Marco Turchi
Předmět:	FOS: Computer and information sciences Computer Science - Computation and Language Computation and Language (cs.CL)
Zdroj:	Web of Science
Popis:	The study of the attention mechanism has sparked interest in many fields, such as language modeling and machine translation. Although its patterns have been exploited to perform different tasks, from neural network understanding to textual alignment, no previous work has analysed the encoder-decoder attention behavior in speech translation (ST) nor used it to improve ST on a specific task. In this paper, we fill this gap by proposing an attention-based policy (EDAtt) for simultaneous ST (SimulST) that is motivated by an analysis of the existing attention relations between audio input and textual output. Its goal is to leverage the encoder-decoder attention scores to guide inference in real time. Results on en->{de, es} show that the EDAtt policy achieves overall better results compared to the SimulST state of the art, especially in terms of computational-aware latency. Accepted to ACL 2023
Databáze:	OpenAIRE
Externí odkaz:	https://explore.openaire.eu/search/publication?articleId=doi_dedup___::317e3a0fddabb2514dc9de91e98fe7f1 https://publons.com/wos-op/publon/56529441/ Zobrazit plný text záznamu