Deaf Speech

Accented speech produced by many individuals who are deaf or significantly hard of hearing, resulting from incomplete acoustic feedback from their own voices. Because deaf speakers cannot fully hear themselves, their speech patterns often differ from those of hearing speakers in predictable ways: phonological errors including sound substitutions and omissions, consonant clustering errors, vowel prolongation, extraneous pauses, irregular syllable duration, and rhythmic inconsistency. The severity varies depending on when the individual lost their hearing, the degree of hearing loss, and speech therapy history. Deaf speech poses a significant accessibility challenge because automatic speech recognition systems are trained predominantly on hearing speech and perform very poorly on deaf speech — with word error rates of 70% or higher — effectively excluding deaf speakers from voice-controlled devices like smart speakers and personal assistants despite the fact that many deaf individuals can and do use spoken language.