Punctuation restoration using transformer models for high-and low-resource languages

T Alam, A Khan, F Alam - Proceedings of the Sixth Workshop on …, 2020 - aclanthology.org
Punctuation restoration is a common post-processing problem for Automatic Speech
Recognition (ASR) systems. It is important to improve the readability of the transcribed text …

A review of bangla natural language processing tasks and the utility of transformer models

F Alam, A Hasan, T Alam, A Khan, J Tajrin… - arXiv preprint arXiv …, 2021 - arxiv.org
Bangla--ranked as the 6th most widely spoken language across the world (https://www.
ethnologue. com/guides/ethnologue200), with 230 million native speakers--is still …

Capitalization and punctuation restoration: a survey

V Păiş, D Tufiş - Artificial Intelligence Review, 2022 - Springer
Ensuring proper punctuation and letter casing is a key pre-processing step towards applying
complex natural language processing algorithms. This is especially significant for textual …

Efficient automatic punctuation restoration using bidirectional transformers with robust inference

M Courtland, A Faulkner… - Proceedings of the 17th …, 2020 - aclanthology.org
Though people rarely speak in complete sentences, punctuation confers many benefits to
the readers of transcribed speech. Unfortunately, most ASR systems do not produce …

Adversarial transfer learning for punctuation restoration

J Yi, J Tao, Y Bai, Z Tian, C Fan - arXiv preprint arXiv:2004.00248, 2020 - arxiv.org
Previous studies demonstrate that word embeddings and part-of-speech (POS) tags are
helpful for punctuation restoration tasks. However, two drawbacks still exist. One is that word …

Automatic punctuation restoration with bert models

A Nagy, B Bial, J Ács - arXiv preprint arXiv:2101.07343, 2021 - arxiv.org
We present an approach for automatic punctuation restoration with BERT models for English
and Hungarian. For English, we conduct our experiments on Ted Talks, a commonly used …

Token-level supervised contrastive learning for punctuation restoration

Q Huang, T Ko, HL Tang, X Liu, B Wu - arXiv preprint arXiv:2107.09099, 2021 - arxiv.org
Punctuation is critical in understanding natural language text. Currently, most automatic
speech recognition (ASR) systems do not generate punctuation, which affects the …

[PDF][PDF] Focal Loss for Punctuation Prediction.

J Yi, J Tao, Z Tian, Y Bai, C Fan - Interspeech, 2020 - interspeech2020.org
Many approaches have been proposed to predict punctuation marks. Previous results
demonstrate that these methods are effective. However, there still exists class imbalance …

Towards better subtitles: A multilingual approach for punctuation restoration of speech transcripts

NM Guerreiro, R Rei, F Batista - Expert Systems with Applications, 2021 - Elsevier
This paper proposes a flexible approach for punctuation prediction that can be used to
produce state-of-the-art results in a multilingual scenario. We have performed experiments …

Boosting punctuation restoration with data generation and reinforcement learning

VD Lai, A Salinas, H Tan, T Bui, Q Tran, S Yoon… - arXiv preprint arXiv …, 2023 - arxiv.org
Punctuation restoration is an important task in automatic speech recognition (ASR) which
aim to restore the syntactic structure of generated ASR texts to improve readability. While …