Dimensions overall score 8.0

GitHub Code Pulse

No public code linked for this paper yet.

• Parakeet-TDT outperforms encoder-decoder and CTC baselines for Huntington's disease speech recognition.result 90%
• HD-specific adaptation reduces WER from 6.99% to 4.95%.result 95%
• The paper proposes a method for using biomarker-based auxiliary supervision for HD speech recognition.method 90%
• Biomarker-based supervision reshapes error behavior in severity-dependent ways rather than uniformly improving WER.result 85%
• The study uses a high-fidelity clinical speech corpus not previously used for end-to-end ASR training.method 90%
• Automatic speech recognition for pathological speech remains underexplored, especially for Huntington's disease.technical 80%
• HD speech presents challenges including irregular timing, unstable phonation, and articulatory distortion.technical 85%
• The study compares multiple ASR families under a unified evaluation including WER and error pattern analysis.method 90%

Competitive landscape

Competitor map is still being generated for this paper. Enable generation or check back soon.

Startup potential card