InterCorrect: Intersection-Aware Correction of Demographic Model Merging for Fair ASR

Key points

• Automatic Speech Recognition (ASR) systems often show uneven performance across demographic groups, and errors can be especially difficult to address for speakers belonging to multiple demographic groups.1
• Starting from a SLAM-ASR-based model, we fine-tune only the connector on demographic-specific subsets and merge the resulting subgroup-adapted connectors into a global model.1
• We then identify critical cross-axis demographic pairs using subgroup WER and task-vector conflict, and apply intersection-specific correction vectors to the global merged model.1
• Experiments on Fair-Speech show that global demographic merging improves overall WER over the base model, while intersection correction provides additional gains for several merging strategies.1

Sources (1)

• [1]InterCorrect: Intersection-Aware Correction of Demographic Model Merging for Fair ASR
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 6, 04:07 PM

  This work studies demographic-aware model merging for fair Speech-LLM-based ASR.

  Automatic Speech Recognition (ASR) systems often show uneven performance across demographic groups, and errors can be especially difficult to address for speakers belonging to multiple demographic groups.

Extractive summary: sentences quoted from the sources.