Samsung’s System for the IWSLT 2019 End-to-End Speech Translation Task

no code implementations • EMNLP (IWSLT) 2019 • Tomasz Potapczyk, Pawel Przybysz, Marcin Chochowski, Artur Szumaczuk

This paper describes the submission to IWSLT 2019 End- to-End speech translation task by Samsung R&D Institute, Poland.

Translation

Short-Term Memory Convolutions

no code implementations • 8 Feb 2023 • Grzegorz Stefański, Krzysztof Arendt, Paweł Daniluk, Bartłomiej Jasik, Artur Szumaczuk

In this work we propose novel method for minimization of inference time latency and memory consumption, called Short-Term Memory Convolution (STMC) and its transposed counterpart.

Acoustic Scene Classification Scene Classification +3