Papers/2609.12122
🧪 Test?View on arXiv

Quantifying Consonant Contributions to Word Intelligibility via Acoustic Masking

Author1, Author2, Author3, Author4, Author5

speech recognitionacoustic analysislanguage processing
2609.12122
Builder Relevance
70%
2h ago

Abstract

This paper presents a scalable method to measure consonant contributions to word intelligibility using acoustic masking.

Reality Card

Core Claim

The study introduces a method to quantify consonant contributions to intelligibility through mask-induced misrecognition rates, validated across multiple languages and ASR architectures.

Method / Result

Consonant contribution scores were derived from the mask-induced misrecognition rate (MMR), showing that phoneme frequency negatively correlates with MMR.

Limitations

The consonant rankings are language-dependent, which may limit the generalizability of the findings across different languages.

Paper to code

Verified implementation resources so builders can test the paper’s claims instead of stopping at the abstract.

No verified implementation link has been attached yet. AIBuzzHub will keep this panel separate from unverified search results.
← Back to all papers