Uloza Virgilijus, Ulozaitė-Stanienė Nora, Petrauskas Tadas, Kregždytė Rima
Department of Otorhinolaryngology, Lithuanian University of Health Sciences, Kaunas, Lithuania.
Department of Otorhinolaryngology, Lithuanian University of Health Sciences, Kaunas, Lithuania.
J Voice. 2023 May;37(3):465.e19-465.e26. doi: 10.1016/j.jvoice.2021.01.025. Epub 2021 Mar 4.
To evaluate the accuracy of Acoustic Voice Quality Index (AVQI) measures obtained from voice recordings simultaneously using oral and smartphone microphones in a sound-proof room, and to compare them with AVQIs obtained from the same smartphone voice recordings with added ambient noise.
A study group of 183 subjects with normal voices (n = 86) and various voice disorders (n = 97) was asked to read aloud a standard text and sustain the vowel /a/. The controlled ambient noise averaged at 29.61 dB SPL was added digitally to the smartphone voice recordings. Repeated measures analysis of variances (ANOVA) with Greenhouse-Geiser correction was used to evaluate AVQI changes within subjects. To evaluate the level of agreement between AVQI measurements obtained from different voice recordings Bland-Altman plots were used.
Repeated measures ANOVA showed that differences among AVQI results obtained from voice recordings done with oral studio microphone, recordings done with a smartphone microphone, and recordings done with a smartphone microphone with added ambient noise were not statistically significant (P = 0.07). No significant systemic differences and acceptable level of random errors in AVQI measurements of voice recordings made with oral and smartphone microphones (including added noise) were revealed.
The AVQI measures obtained from smartphone microphones voice recordings with experimentally added ambient noise revealed an acceptable agreement with results of oral microphone recordings, thus suggesting the suitability of smartphone microphone recordings performed even in the presence of acceptable ambient noise for estimation of AVQI.
评估在隔音室内同时使用口腔麦克风和智能手机麦克风从语音录音中获得的声学语音质量指数（AVQI）测量的准确性，并将其与从添加了环境噪声的相同智能手机语音录音中获得的AVQI进行比较。
要求一个由183名具有正常嗓音（n = 86）和各种嗓音障碍（n = 97）的受试者组成的研究组大声朗读一篇标准文本并持续发元音/a/。将平均为29.61 dB SPL的受控环境噪声数字添加到智能手机语音录音中。使用带有Greenhouse-Geiser校正的重复测量方差分析（ANOVA）来评估受试者内的AVQI变化。为了评估从不同语音录音获得的AVQI测量之间的一致性水平，使用了Bland-Altman图。
重复测量方差分析表明，使用口腔录音室麦克风进行的语音录音、使用智能手机麦克风进行的录音以及添加了环境噪声的智能手机麦克风进行的录音所得到的AVQI结果之间的差异无统计学意义（P = 0.07）。在使用口腔和智能手机麦克风（包括添加噪声的情况）进行的语音录音的AVQI测量中，未发现显著的系统差异和可接受水平的随机误差。
从添加了实验性环境噪声的智能手机麦克风语音录音中获得的AVQI测量结果与口腔麦克风录音结果显示出可接受的一致性，因此表明即使在存在可接受的环境噪声的情况下，智能手机麦克风录音也适用于AVQI的估计。
