Locality enhanced dynamic biasing and sampling strategies for contextual ASR

Autor:	Jalal, Md Asif, Parada, Pablo Peso, Pavlidis, George, Moschopoulos, Vasileios, Saravanan, Karthikeyan, Kontoulis, Chrysovalantis-Giorgos, Zhang, Jisi, Drosou, Anastasios, Lee, Gil Ho, Lee, Jungin, Jung, Seokyeong
Rok vydání:	2024
Předmět:	Electrical Engineering and Systems Science - Audio and Speech Processing Computer Science - Computation and Language Computer Science - Sound
Druh dokumentu:	Working Paper
Popis:	Automatic Speech Recognition (ASR) still face challenges when recognizing time-variant rare-phrases. Contextual biasing (CB) modules bias ASR model towards such contextually-relevant phrases. During training, a list of biasing phrases are selected from a large pool of phrases following a sampling strategy. In this work we firstly analyse different sampling strategies to provide insights into the training of CB for ASR with correlation plots between the bias embeddings among various training stages. Secondly, we introduce a neighbourhood attention (NA) that localizes self attention (SA) to the nearest neighbouring frames to further refine the CB output. The results show that this proposed approach provides on average a 25.84% relative WER improvement on LibriSpeech sets and rare-word evaluation compared to the baseline. Comment: Accepted for IEEE ASRU 2023
Databáze:	arXiv
Externí odkaz:	http://arxiv.org/abs/2401.13146 Zobrazit plný text záznamu View this record from Arxiv