用球谐变换实现跨麦克风阵列的轻量语音识别
Lightweight and Robust Multi-Channel End-to-End Speech Recognition with Spherical Harmonic Transform
- 用球谐变换将麦克风信号转为与阵列结构无关的系数
- 在异构阵列上平均字错误率39.26%，计算量减少97.1%
- 适合部署在资源受限设备，尤其多阵列场景
本文提出SHTNet，一种基于球谐变换（SHT）的轻量级多通道自动语音识别框架，旨在解决跨阵列泛化难题。首先，基于SHT的空间声场分解将麦克风信号转换为几何不变的球谐系数，使信号处理脱离阵列结构依赖。其次，空间-谱注意力融合网络（SSAFN）结合坐标感知的空间建模、优化的自注意力通道组合及谱域噪声抑制，无需传统波束成形。第三，随机球谐变换（Rand-SHT）训练通过随机选择通道和重构阵列几何提升鲁棒性。系统在Aishell-4、Alimeeting和XMOS等数据集上，面对圆形、方形及双耳等异构阵列，平均字错误率达39.26%，计算量仅为传统神经波束成形的2.9%。
原文摘要 · Abstract (English)
This paper presents SHTNet, a lightweight spherical harmonic transform (SHT) based framework, which is designed to address cross-array generalization challenges in multi-channel automatic speech recognition (ASR) through three key innovations. First, SHT based spatial sound field decomposition converts microphone signals into geometry-invariant spherical harmonic coefficients, isolating signal processing from array geometry. Second, the Spatio-Spectral Attention Fusion Network (SSAFN) combines coordinate-aware spatial modeling, refined self-attention channel combinator, and spectral noise suppression without conventional beamforming. Third, Rand-SHT training enhances robustness through random channel selection and array geometry reconstruction. The system achieves 39.26\% average CER across heterogeneous arrays (e.g., circular, square, and binaural) on datasets including Aishell-4, Alimeeting, and XMOS, with 97.1\% fewer computations than conventional neural beamformers.
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。