Efficient Ensemble for Multimodal Punctuation Restoration using Time-Delay Neural Network

Autor:	Liu, Xing Yi, Beigi, Homayoon
Rok vydání:	2023
Předmět:	Computer Science - Computation and Language Computer Science - Human-Computer Interaction Computer Science - Machine Learning Computer Science - Sound Electrical Engineering and Systems Science - Audio and Speech Processing
Zdroj:	2024 18th International Conference on Ubiquitous Information Management and Communication (IMCOM), 2024, pp. 1-6
Druh dokumentu:	Working Paper
DOI:	10.1109/IMCOM60618.2024.10418445
Popis:	Punctuation restoration plays an essential role in the post-processing procedure of automatic speech recognition, but model efficiency is a key requirement for this task. To that end, we present EfficientPunct, an ensemble method with a multimodal time-delay neural network that outperforms the current best model by 1.0 F1 points, using less than a tenth of its inference network parameters. We streamline a speech recognizer to efficiently output hidden layer acoustic embeddings for punctuation restoration, as well as BERT to extract meaningful text embeddings. By using forced alignment and temporal convolutions, we eliminate the need for attention-based fusion, greatly increasing computational efficiency and raising performance. EfficientPunct sets a new state of the art with an ensemble that weights BERT's purely language-based predictions slightly more than the multimodal network's predictions. Our code is available at https://github.com/lxy-peter/EfficientPunct. Comment: 6 pages, 1 figure, 5 tables, paper at IMCOM 2024, technical report at Recognition Technologies, Inc
Databáze:	arXiv
Externí odkaz:	http://arxiv.org/abs/2302.13376 Zobrazit plný text záznamu View this record from Arxiv