针对口吃者语音唤醒难题，提出端到端双滤波系统，显著降低误唤醒率。
Optimizing Dysarthria Wake-Up Word Spotting: An End-to-End Approach for SLT 2024 LRDWWS Challenge
- 采用多任务微调的双分支data2vec2模型，统一建模语音识别与唤醒词检测
- 在测试集上实现0.321%误唤醒率、0.5%漏唤醒率，总分0.821%
- 适合残障语音交互、低资源口吃语音处理场景
语音已成为各类应用中广泛采用的用户界面。然而，对于患有构音障碍的人群，其语音固有的不稳定性带来了显著挑战。本文针对SLT 2024低资源构音障碍语音唤醒词检测挑战赛，提出一种基于预训练的端到端双滤波构音障碍语音唤醒系统（PD-DWS）。该系统从音频建模与双滤波策略两方面提升性能：在音频建模方面，提出基于预训练data2vec2（d2v2）的双分支2branch-d2v2模型，通过统一的多任务微调范式，同时建模自动语音识别（ASR）与语音唤醒词检测（WWS）任务；此外，引入双滤波策略，在保持相同漏唤醒率（FRR）的前提下，有效降低误唤醒率（FAR）。实验结果表明，所提PD-DWS系统在测试集B上达到FAR为0.00321、FRR为0.005，总分为0.00821，获得挑战赛第一名。
原文摘要 · Abstract (English)
Speech has emerged as a widely embraced user interface across diverse applications. However, for individuals with dysarthria, the inherent variability in their speech poses significant challenges. This paper presents an end-to-end Pretrain-based Dual-filter Dysarthria Wake-up word Spotting (PD-DWS) system for the SLT 2024 Low-Resource Dysarthria Wake-Up Word Spotting Challenge. Specifically, our system improves performance from two key perspectives: audio modeling and dual-filter strategy. For audio modeling, we propose an innovative 2branch-d2v2 model based on the pre-trained data2vec2 (d2v2), which can simultaneously model automatic speech recognition (ASR) and wake-up word spotting (WWS) tasks through a unified multi-task finetuning paradigm. Additionally, a dual-filter strategy is introduced to reduce the false accept rate (FAR) while maintaining the same false reject rate (FRR). Experimental results demonstrate that our PD-DWS system achieves an FAR of 0.00321 and an FRR of 0.005, with a total score of 0.00821 on the test-B eval set, securing first place in the challenge.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。