Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

GLoRIA: Gated Low-Rank Interpretable Adaptation for Dialectal ASR

Pouya Mehralian, Melissa Farasyn, Anne Breitbarth, Anne-Sophie Ghyselen, Hugo Van hamme

TL;DR

Results show metadata-gated low-rank adaptation is an effective, interpretable, and efficient solution for dialectal ASR.

Abstract

Automatic Speech Recognition (ASR) in dialect-heavy settings remains challenging due to strong regional variation and limited labeled data. We propose GLoRIA, a parameter-efficient adaptation framework that leverages metadata (e.g., coordinates) to modulate low-rank updates in a pre-trained encoder. GLoRIA injects low-rank matrices into each feed-forward layer, with a gating MLP determining the non-negative contribution of each LoRA rank-1 component based on location metadata. On the GCND corpus, GLoRIA outperforms geo-conditioned full fine-tuning, LoRA, and both dialect-specific and unified full fine-tuning, achieving state-of-the-art word error rates while updating under 10% of parameters. GLoRIA also generalizes well to unseen dialects, including in extrapolation scenarios, and enables interpretable adaptation patterns that can be visualized geospatially. These results show metadata-gated low-rank adaptation is an effective, interpretable, and efficient solution for dialectal ASR.

GLoRIA: Gated Low-Rank Interpretable Adaptation for Dialectal ASR

TL;DR

Results show metadata-gated low-rank adaptation is an effective, interpretable, and efficient solution for dialectal ASR.

Abstract

Automatic Speech Recognition (ASR) in dialect-heavy settings remains challenging due to strong regional variation and limited labeled data. We propose GLoRIA, a parameter-efficient adaptation framework that leverages metadata (e.g., coordinates) to modulate low-rank updates in a pre-trained encoder. GLoRIA injects low-rank matrices into each feed-forward layer, with a gating MLP determining the non-negative contribution of each LoRA rank-1 component based on location metadata. On the GCND corpus, GLoRIA outperforms geo-conditioned full fine-tuning, LoRA, and both dialect-specific and unified full fine-tuning, achieving state-of-the-art word error rates while updating under 10% of parameters. GLoRIA also generalizes well to unseen dialects, including in extrapolation scenarios, and enables interpretable adaptation patterns that can be visualized geospatially. These results show metadata-gated low-rank adaptation is an effective, interpretable, and efficient solution for dialectal ASR.
Paper Structure (28 sections, 4 equations, 2 figures, 2 tables)

This paper contains 28 sections, 4 equations, 2 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Dialect-Specific Approaches
  4. Metadata-Aware Unified Models
  5. Continuous Geo-Conditioning
  6. Dataset
  7. Pretrained Model
  8. Methods
  9. LoRA Recap and GLoRIA
  10. Regularization Losses
  11. Training Objective and Procedure
  12. Experimental Setup
  13. Dataset and Dialect Selection
  14. Model Configuration
  15. Training Configuration
  16. ...and 13 more sections

Figures (2)

  • Figure 1: Geographical distribution of activation for four NMF-derived adaptation components. Each point represents a location; color intensity reflects the degree of adaptation component usage at that location. The emergent spatial patterns closely correspond to known dialect regions. The patterns shown correspond to the Frans-Vlaams, Limburgs, Oost-Vlaams and Antwerp regions, from left to right.
  • Figure 2: Clustered heatmap of NMF component activations averaged by dialect region. Regions that are geographically and acoustically closer cluster together, indicating interpretable adaptation behavior.