ASR Personalization in Pathological Voice: Accuracy, Efficiency, and Duration Effects
Automatic Speech recognition (ASR) struggles with Pathological speech due to persistent speaker-specific mismatch.
ProofPaper ↗
Key points
- In clinical and accessibility scenarios, only limited target-speaker data is typically available, making full model retraining impractical.
- We present a component-level analysis of LoRA adaptation in Whisper, investigating how adapter placement interacts with the model size, training efficiency and utterance-duration in single-speaker settings.
- Our experimental comparative study shows that while scaling reduces errors, it does not eliminate mismatch.
- These findings highlight the importance of adaptation placement and duration-dependent evaluation, under challenging, data-scarce personalization scenarios.
Sources (1)
- [1]ASR Personalization in Pathological Voice: Accuracy, Efficiency, and Duration EffectsarXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 17, 09:05 AM
Automatic Speech recognition (ASR) struggles with Pathological speech due to persistent speaker-specific mismatch.
In clinical and accessibility scenarios, only limited target-speaker data is typically available, making full model retraining impractical.
Extractive summary: sentences quoted from the sources.
Before this
- Oct 16, 2026Acceptance Is Not Speed: Length Inflation in Training-Free Lossy Speculative Decoding
- Oct 16, 2026Momentum: Persistent Dynamics for Video World Models
- Oct 16, 2026SpikeQ: Spike-Guided Post-Training Quantization for Spiking Neural Networks
- Oct 15, 2026Granite Switch 4.1: Chat, Decisions, and Speech from One Endpoint
- Oct 15, 2026Accelerating the search for antimicrobial peptides with Google Antigravity
- Oct 15, 2026FuseAdapt: Parameter-Efficient Multimodal Fusion for Semantic Segmentation with Missing Modalities