توضح هذه الورقة تقديم نظام الترجمة من Niutrans End-tou-end الكلام للمهمة غير المتصلة IWSLT 2021، والتي تترجم من الصوت الإنجليزي إلى النص الألماني مباشرة دون نسخ متوسط.نحن نستخدم الهندسة المعمارية النموذجية القائمة على المحولات وتعزيزها عن طريق مطابقة، ترميز الموضع النسبي، والترميز الصوتية والترميز النصي مكدسة.لزيادة بيانات التدريب، يتم ترجم نسخ اللغة الإنجليزية إلى الترجمات الألمانية.أخيرا، نحن نوظف فك تشفير الفرقة لدمج التنبؤات من عدة نماذج مدربة مع مجموعات البيانات المختلفة.الجمع بين هذه التقنيات، نحقق 33.84 نقطة بلو على مجموعة اختبار EN-DE MUST-C، والتي تظهر الإمكانات الهائلة لنموذج نهاية إلى نهاية.
This paper describes the submission of the NiuTrans end-to-end speech translation system for the IWSLT 2021 offline task, which translates from the English audio to German text directly without intermediate transcription. We use the Transformer-based model architecture and enhance it by Conformer, relative position encoding, and stacked acoustic and textual encoding. To augment the training data, the English transcriptions are translated to German translations. Finally, we employ ensemble decoding to integrate the predictions from several models trained with the different datasets. Combining these techniques, we achieve 33.84 BLEU points on the MuST-C En-De test set, which shows the enormous potential of the end-to-end model.
References used
https://aclanthology.org/
This paper describes the ESPnet-ST group's IWSLT 2021 submission in the offline speech translation track. This year we made various efforts on training data, architecture, and audio segmentation. On the data side, we investigated sequence-level knowl
This paper describes KIT'submission to the IWSLT 2021 Offline Speech Translation Task. We describe a system in both cascaded condition and end-to-end condition. In the cascaded condition, we investigated different end-to-end architectures for the spe
In this paper, we describe Zhejiang University's submission to the IWSLT2021 Multilingual Speech Translation Task. This task focuses on speech translation (ST) research across many non-English source languages. Participants can decide whether to work
This paper describes the submission to the IWSLT 2021 Low-Resource Speech Translation Shared Task by IMS team. We utilize state-of-the-art models combined with several data augmentation, multi-task and transfer learning approaches for the automatic s
This paper describes Maastricht University's participation in the IWSLT 2021 multilingual speech translation track. The task in this track is to build multilingual speech translation systems in supervised and zero-shot directions. Our primary system