SIMD-size aware weight regularization for fast neural vocoding on CPU
This paper proposes weight regularization for a faster neural vocoder. Pruning time-consuming DNN modules is a promising way to realize a real-time vocoder on a CPU (e.g. WaveRNN, LPCNet). Regularization that encourages sparsity is also effective in avoiding the quality degradation created by pruning. However, the orders of weight matrices must be contiguous in SIMD size for fast vocoding. To ensure this order, we propose explicit SIMD size aware regularization. Our proposed method reshapes a weight matrix into a tensor so that the weights are aligned by group size in advance, and then computes the group Lasso-like regularization loss. Experiments on 70% sparse subband WaveRNN show that pruning in conventional Lasso and column-wise group Lasso degrades the synthetic speech's naturalness. The vocoder with proposed regularization 1) achieves comparable naturalness to that without pruning and 2) performs meaningfully faster than other conventional vocoders using regularization.
Code (0)
등록된 구현이 없습니다.
Tasks
CPUMethods 이 논문이 사용한 방법론
Similar Papers 제목 키워드 기반
VibeVoice-ASR-BitNet Technical Report
We present VibeVoice-ASR-BitNet, a compressed variant of VibeVoice-ASR optimized for real-time inference on edge CPUs. We apply heterogeneous quantization tailored to the computational characteristics of each stage: the …
ExaGEMM: Exploration Framework for CPU-Driven ML Inference via Associative In-Register Computing for Low-Bit GEMM
Low-bit GEMM is increasingly central to efficient ML inference, yet very-low-bit execution remains a poor fit for conventional CPUs. Practical deployment spans fragmented regimes-from 1/2/4-bit weights to varying activat…
NoMAD-Attention: Efficient LLM Inference on CPUs Through Multiply-add-free Attention
Large language model inference on Central Processing Units (CPU) is challenging due to the vast quantities of expensive Multiply-Add (MAD) matrix operations in the attention computations. In this paper, we argue that the…
16kCPULanguage ModelingLanguage Modelling+1Decoding billions of integers per second through vectorization
In many important applications -- such as search engines and relational database systems -- data is stored in the form of arrays of integers. Encoding and, most importantly, decoding of these arrays consumes considerable…
CPUISimDL: Importance Sampling-Driven Acceleration of Fault Injection Simulations for Evaluating the Robustness of Deep Learning
Deep Learning (DL) systems have proliferated in many applications, requiring specialized hardware accelerators and chips. In the nano-era, devices have become increasingly more susceptible to permanent and transient faul…