We propose a novel neural model compression strategy combining data augmentation, knowledge transfer, pruning, and quantization for device-robust acoustic scene classification (ASC). Specifically, we tackle the ASC task in a low-resource environment leveraging a recently proposed advanced neural network pruning mechanism, namely Lottery Ticket Hypothesis (LTH), to find a sub-network neural model associated with a small amount non-zero model parameters. The effectiveness of LTH for low-complexity acoustic modeling is assessed by investigating various data augmentation and compression schemes, and we report an efficient joint framework for low-complexity multi-device ASC, called Acoustic Lottery. Acoustic Lottery could compress an ASC model over $1/10^{4}$ and attain a superior performance (validation accuracy of 74.01% and Log loss of 0.76) compared to its not compressed seed model. All results reported in this work are based on a joint effort of four groups, namely GT-USTC-UKE-Tencent, aiming to address the "Low-Complexity Acoustic Scene Classification (ASC) with Multiple Devices" in the DCASE 2021 Challenge Task 1a.
翻译:我们提出了一个新颖的神经模型压缩战略,其中结合了数据增强、知识转移、剪裁和测量设备-气压声学场景分类(ASC ) 。 具体地说,我们在低资源环境中利用最近提出的先进神经网络运行机制(即Lottery Ticket Hypothes(LTH)),在低资源环境中处理ASC任务,以便找到一个小量非零模型参数相关的子网络神经模型。通过调查各种数据增强和压缩计划,评估低兼容性声学模型LTH的效力。我们报告了一个低兼容性多功能低联合框架,称为Acoucistic Lotry。 听觉彩票可以压缩1/10+4美元以上的ASC模型,并取得优异性性性表现(校验率精确率为74.01%,日志损失为0.76美元,与其非压缩种子模型相比。 这项工作上报告的所有结果都基于四组的联合努力,即GT-USTC-UKE-ENent,目的是解决“D-Com-comliclicity Astable 1 Clistristrational Acal”的DC-Sex Acal 20-CA 20-Sex Astistristristration。