arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2207.03331eess.AScs.LG
收藏

基于知识蒸馏的低资源低占用唤醒词检测

Low-resource Low-footprint Wake-word Detection using Knowledge Distillation

M Health Information Systems(3M健康信息系统公司)

Arindam Ghosh, Mark Fuhs, Deblin Bagchi, Bahman Farahani, Monika Woszczyna

摘要:

随着虚拟助手变得越来越多样化和专业化,对应用程序或品牌特定唤醒词的需求也越来越大。然而,通常用于训练唤醒词检测器的唤醒词特定数据集的创建成本很高。在本文中,我们探讨了两种技术,利用声学建模数据的大词汇量语音识别,以提高一个专门建造的唤醒词检测器:迁移学习和知识蒸馏。我们还探讨了这些技术如何与时间同步的训练目标相互作用,以改善检测延迟。实验上的开源“嘿Snips”数据集和更具挑战性的内部远场数据集。使用电话同步目标和从大型声学模型中提取的知识,我们能够提高两个数据集的数据集大小的准确性,同时减少延迟。

英文摘要:

As virtual assistants have become more diverse and specialized, so has the demand for application or brand-specific wake words. However, the wake-word-specific datasets typically used to train wake-word detectors are costly to create. In this paper, we explore two techniques to leverage acoustic modeling data for large-vocabulary speech recognition to improve a purpose-built wake-word detector: transfer learning and knowledge distillation. We also explore how these techniques interact with time-synchronous training targets to improve detection latency. Experiments are presented on the open-source "Hey Snips" dataset and a more challenging in-house far-field dataset. Using phone-synchronous targets and knowledge distillation from a large acoustic model, we are able to improve accuracy across dataset sizes for both datasets while reducing latency.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑