Chunk Size Scheduling for Optimizing the Quality-Latency Trade-off in Simultaneous Speech Translation

Iqbal Pahlevi Amin, Haotian Tan, Kurniawati Azizah, Sakriani Sakti

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

This study addresses the quality-latency trade-off in simultaneous speech translation (SimulST) by proposing dynamic chunk size scheduling during evaluation. Traditional approaches wait for full speech before the translation begin, leading to delays. SimulST aims for minimal delay and by starting the translation incrementally. The proposed method schedules changes to chunk size during evaluation to gradually reduce predefined chunk size, aiming to maintain trans-lation quality while reducing latency. Evaluation using the MuST-C v2.0 tst-COMMON dataset shows promising results, especially in high latency regimes for English-to-German translations.

Original languageEnglish
Title of host publication2024 27th Conference on the Oriental COCOSDA International Committee for the Co-Ordination and Standardisation of Speech Databases and Assessment Techniques, O-COCOSDA 2024 - Proceedings
EditorsMing-Hsiang Su, Jui-Feng Yeh, Yuan-Fu Liao, Chi-Chun Lee, Yu Taso
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9798331506032
DOIs
Publication statusPublished - 2024
Event27th Conference on the Oriental COCOSDA International Committee for the Co-Ordination and Standardisation of Speech Databases and Assessment Techniques, O-COCOSDA 2024 - Hsinchu, Taiwan, Province of China
Duration: 17 Oct 202419 Oct 2024

Publication series

Name2024 27th Conference on the Oriental COCOSDA International Committee for the Co-Ordination and Standardisation of Speech Databases and Assessment Techniques, O-COCOSDA 2024 - Proceedings

Conference

Conference27th Conference on the Oriental COCOSDA International Committee for the Co-Ordination and Standardisation of Speech Databases and Assessment Techniques, O-COCOSDA 2024
Country/TerritoryTaiwan, Province of China
CityHsinchu
Period17/10/2419/10/24

Keywords

  • chunk size scheduling
  • simultaneous speech translation
  • speech-to-text translation

Fingerprint

Dive into the research topics of 'Chunk Size Scheduling for Optimizing the Quality-Latency Trade-off in Simultaneous Speech Translation'. Together they form a unique fingerprint.

Cite this