Choosing a PEFT Variant for Per-Patient Dysarthric ASR: A Single-Speaker Case Study on Two ASR Bases
Per-patient adapters are the preferred production architecture for dysarthric automatic speech recognition (ASR), yet parameter-efficient fine-tuning (PEFT) variants have not been compared in the speaker-dependent, per-patient regime. We present a single-speaker case study comparing seven LoRA-family methods (LoRA, QLoRA, AdaLoRA, DoRA, LoHA, VeRA, VB-LoRA) on two production bases (Whisper-large-v3 with Hungarian fine-tuning, and a multilingual Qwen3-ASR-1.7B checkpoint) for one post-stroke Hungarian male speaker (S1, 409 utterances; severe dysarthria on auditory-perceptual clinical assessment
Lineage graph
Paper → model → repo connections mined from source citations (Tier-1 exact match).
Why these links exist
Every edge carries a method, confidence, and the source snippet that justified it — so bad links are debuggable.
- FuzzySimilar title/name (fuzzy) · 59%pyannote/speaker-diarization-3.1 →
“Fuzzy title match (0.73): “Choosing a PEFT Variant for Per-Patient Dysarthric ASR: A Si” ≈ “pyannote/speaker-diarization-3.1””
- PossiblePossibly related (embedding) · 47%Granite Speech 5.0 Turbo CTC: Extremely Fast and Accurate Transcription →
- PossiblePossibly related (embedding) · 47%Measuring benchmark optimization in speech recognition →
- PossiblePossibly related (embedding) · 47%The bottleneck for meeting transcription tools isn't accurate anymore, it's speaker attribution →
- PossiblePossibly related (embedding) · 46%Launch HN: Speko (YC S26) – OpenRouter for Voice AI →
- LinkedLinked via arxiv author · 85%Bernard Muller →
“Choosing a PEFT Variant for Per-Patient Dysarthric ASR: A Single-Speaker Case Study on Two ASR Bases”
- LinkedLinked via arxiv author · 85%László Tóth →
“Choosing a PEFT Variant for Per-Patient Dysarthric ASR: A Single-Speaker Case Study on Two ASR Bases”
- LinkedLinked via arxiv author · 85%LaVonne Roberts →
“Choosing a PEFT Variant for Per-Patient Dysarthric ASR: A Single-Speaker Case Study on Two ASR Bases”