Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

GC-LoRA: Gated Convolutional LoRA for Parameter-Efficient Acoustic Adaptation

About

Transformer-based Speech Foundation Models excel in most Automatic Speech Recognition tasks but often suffer performance degradation when applied to domains with mismatched acoustic characteristics. While Parameter Efficient Fine-Tuning (PEFT) methods, such as Low-Rank Adaptation (LoRA), adjust global attention, they lack the local context modeling crucial for capturing domain-specific variations. We propose GC-LoRA, a novel adapter architecture that injects Conformer-style local convolutional processing into pretrained Transformer encoders. By integrating a lightweight adapter to encoder attention output projections, our method efficiently captures local acoustic dependencies without disrupting pretrained global representations. Experiments across diverse datasets (acoustically-degraded, bandlimited, dialectal, child) demonstrate the efficacy of our approach, achieving Word Error Rate (WER) reductions of up to 10.9% compared to baselines while adding minimal trainable parameters.

Natarajan Balaji Shankar, Zilai Wang, Kaiyuan Zhang, Mohan Shi, Abeer Alwan• 2026

Related benchmarks

TaskDatasetResultRank
Automatic Speech RecognitionMyST (test)
WER8.6
52
Automatic Speech RecognitionAMI (test)
Word Error Rate11.5
38
Automatic Speech RecognitionCORAAL (test)
WER9.7
23
Speech RecognitionSwitchboard (SWBD) (test)
WER6.3
14
Showing 4 of 4 rows

Other info

Follow for update