Acoustic methods for analysing breathy and whispery voices: a systematic review

Voice quality (VQ) is frequently analysed in phonetic and phonological research. Recently, there has been motivation to assess VQ more ‘objectively’ using acoustics. Our systematic review revealed a notable research focus on pathological speakers, with 48% of excluded studies being rejected for investigating pathological populations. Among studies involving non-pathological speakers, the analysed speech material is often restricted to sustained vowel productions. Therefore, the reliability of acoustic techniques when analysisng more naturalistic speech and non-pathological speakers remains unclear. We present a systematic literature review of acoustic methods used to categorise breathy and whispery VQ in non-pathological speakers. The literature was surveyed using four databases (ProQuest, PubMed, SCOPUS and Web of Science) and ICPhS proceedings between 1999-2023. The selection criteria included peer-reviewed articles conducting an acoustic analysis of breathy and whispery voices for vocally healthy speakers. Initial searches yielded 754 papers. Once filtered 21 papers remained. The results reveal some consistency in the main acoustic parameters for breathy VQ (higher spectral tilt and lower HNR/CPP). Whispery voice was only addressed in two studies meaning no trends were observed. We conclude that there remain inconsistencies in methods and findings, meaning we cannot identify an agreed ‘standard’ approach generally applicable to non-pathological speakers.