AP-GRPO: Anchor-Gated Phonetic Alignment with Policy Optimization for Pathological Speech Reconstruction

2026-06-13

AI Summary

This work addresses the challenges of acoustic distortion and linguistic fragmentation commonly observed in the speech of individuals with neurodegenerative and neuromotor disorders. To tackle these issues, the authors propose AP-GRPO, a novel method that identifies intelligible speech segments as semantic anchors and introduces a dual-reward mechanism—comprising anchor preservation and inter-anchor phoneme alignment—to guide speech language models toward semantically consistent reconstruction. The approach innovatively integrates an anchor-gating strategy with disease-adaptive reinforcement learning constraints within a Group Relative Policy Optimization (GRPO) framework, enabling interpretable and pathology-specific speech restoration. Experimental results demonstrate that AP-GRPO significantly improves reconstruction fidelity across four distinct neurological conditions and automatically modulates anchor constraint strength to reflect the unique speech degradation patterns associated with each disorder.

Abstract

Pathological speech from patients with neurodegenerative and neuromotor disorders is often acoustically distorted and linguistically fragmented, making pathological speech reconstruction necessary to recover intended textual content from distorted and incomplete speech recordings. Crucially, such recordings are rarely uniformly degraded: some words or short phrases remain reliable and can serve as audible anchors for reconstructing the corrupted surrounding content. We introduce Anchor-gated Phonetic Group Relative Policy Optimization (AP-GRPO), a GRPO framework with phonetic reward that aligns speech language models (SLMs) through audible-anchor preservation and inter-anchor phonetic compatibility to the original speech signal. AP-GRPO consists of: (i) an anchor-gated reward that matches reliable audible anchors in clear regions; and (ii) an inter-anchor phonetic alignment reward that evaluates whether recovered contents are phonetically supported by the corresponding corrupted inter-anchor speech span. Across four disease conditions, AP-GRPO improves faithful speech reconstruction, and the learned anchor constraint automatically adapts to each condition and thus reveals interpretable disease-specific profiles: conditions with severe articulatory degradation require stronger anchor enforcement, whereas milder impairment or linguistically impaired conditions rely more on phonetic alignment for inter-anchor recovery.

Problem

• pathological speech reconstruction
• audible anchors
• phonetic alignment
• speech language models
• neurodegenerative disorders

Innovation

• anchor-gated alignment
• phonetic reward
• pathological speech reconstruction
• speech language models
• policy optimization