Automated Assessment of Hypernasality in Motor Speech Disorders Using Nasal Cognate Distinctiveness and Nasal Posterior Probability

The study evaluated two automated nasality measurement methods - Nasal Cognate Distinctiveness (NCD) and nasal posterior probability (NPP) - in 561 patients, including 374 neurologically typical controls and 187 individuals with motor speech disorders. Researchers analyzed sentence-level speech recordings and compared results with speech-language pathologist ratings. Both methods showed consistent changes with increasing hypernasality severity: NCD decreased while NPP increased. NCD and NPP demonstrated comparable ability to distinguish between absent and present hypernasality with AUC values of 0.796 and 0.730, respectively. The NPP method captured nasalization information in both consonants and vowels with identical classification performance (AUC 0.853). Test-retest reliability was good with minimal changes in the control group. Both methods provide objective markers of hypernasality in patients with motor speech disorders.