SSVD-O: Parameter-Efficient Fine-Tuning with Structured SVD for Speech Recognition
Parameter-efficient fine-tuning (PEFT) is a scalable approach for adapting large speech foundation models to new domains. While methods such as LoRA and its state-of-the-art variants reduce adaptation costs, they typically allocate parameters uniformly across model subspaces, which limits their efficiency and scalability in speech applications. Building on our prior work, this paper introduces SSVD-Outer (SSVD-O), an extension of the structured SVD-guided (SSVD) fine-tuning method. SSVD-O combines input acoustic feature space-associated inner transformations with output semantic feature space-associated outer transformations to enable scalable and balanced adaptation. We conduct the first systematic analysis of parameter budget allocation across model subspaces in PEFT for automatic speech recognition (ASR), and investigate the trade-off between learning and forgetting under constrained resources. SSVD-O is benchmarked against LoRA, DoRA, PiSSA, and SSVD on domain-shifted ASR tasks, including child speech and regional accents, across model scales from 0.1B to 2B within the ESPnet framework. Experimental results show that SSVD-O consistently narrows the performance gap to full fine-tuning while improving generalization and mitigating catastrophic forgetting.
Code (0)
등록된 구현이 없습니다.
Tasks
parameter-efficient fine-tuningSpeech RecognitionSimilar Papers 제목 키워드 기반
SSVD: Structured SVD for Parameter-Efficient Fine-Tuning and Benchmarking under Domain Shift in ASR
Parameter-efficient fine-tuning (PEFT) has emerged as a scalable solution for adapting large foundation models. While low-rank adaptation (LoRA) is widely used in speech applications, its state-of-the-art variants, e.g.,…
parameter-efficient fine-tuningSpeech RecognitionDomain AdaptationSMSSVD - SubMatrix Selection Singular Value Decomposition
High throughput biomedical measurements normally capture multiple overlaid biologically relevant signals and often also signals representing different types of technical artefacts like e.g. batch effects. Signal identifi…
Dimensionality ReductionSingle Shot Video Object Detector
Single shot detectors that are potentially faster and simpler than two-stage detectors tend to be more applicable to object detection in videos. Nevertheless, the extension of such object detectors from image to video is…
GPUObjectobject-detectionObject Detection+1Parameter-Efficient Fine-Tuning for Low-Resource Languages: A Comparative Study of LLMs for Bengali Hate Speech Detection
Bengali social media platforms have witnessed a sharp increase in hate speech, disproportionately affecting women and adolescents. While datasets such as BD-SHS provide a basis for structured evaluation, most prior appro…
parameter-efficient fine-tuningHate Speech DetectionResSVD: Residual Compensated SVD for Large Language Model Compression
Large language models (LLMs) have demonstrated impressive capabilities in a wide range of downstream natural language processing tasks. Nevertheless, their considerable sizes and memory demands hinder practical deploymen…
Language ModelingLanguage ModellingLarge Language ModelModel Compression