Search

Prediction of the apnea-hypopnea index (AHI) from breathing sounds during sleep could be used to prescreen for obstructive sleep apnea (OSA). In addition, the oxygen desaturation index (ODI) is a known risk factor for developing cardiovascular disease in OSA patients. This study focused on estimation of ODI from a noncontact manner from sleep breathing sounds.Retrospective study.Patients who visited the sleep center due to snoring or sleep apnea underwent polysomnography in lab overnight. Sound recordings were made during polysomnography using a microphone. After noise reduction, the sound data were segmented into 5 seconds windows and features were extracted. Binary classification and regression analyses were performed to estimate the ODI during sleep (model 1). This was re-tested after inclusion of body mass index (BMI) and age as additional features (model 2: BMI only, model 3: BMI and age).We included 116 patients. The mean age and AHI of all patients were 50.4 ± 16.7 years and 23.0 ± 24.0 events/hr. In binary classification, for ODI cutoff values of 5, 15, and 30 events/hr, the areas under the curve were 0.88, 0.93, 0.91, respectively, and accuracies were 85.34, 86.21, and 87.07, respectively. In regression analysis, the correlation coefficient and mean absolute error were 0.80 and 9.60 events/hr, respectively. In models 2 and 3, the correlation coefficient and mean absolute error were 0.82, 9.44 events/hr and 0.81, 9.6 events/hr, respectively.Prediction of ODI from sleep sound seems to be feasible. Additional clinical feature such as BMI may increase overall predictability.4 Laryngoscope, 132:901-905, 2022.

PurposeThe authors aim to explore whether and how variations on pitch levels of background music in audiovisual commercials affect consumers' judgments of the competence of the advertised brands and for which group of consumers such changes in ad backgrounds are more influential.Design/methodology/approachConsumers are presented with an audiovisual advertisement in which the pitch of background music is lowered or raised. They are subsequently asked to evaluate the music and traits of the advertised brand and indicate their predisposed styles of thinking.FindingsConsumers tend to judge a brand in an audiovisual commercial as possessing a higher level of competence traits when the brand is accompanied by lower-pitched (vs higher-pitched) background music, which is mediated by levels of powerfulness they perceive from the background music. Consumers with holistic (vs analytic) thinking styles, who are known to devote more focused attention to background information, tend to be more (vs less) susceptible to the changes in pitch.Research limitations/implicationsThe current research approaches thinking styles as predisposed individual differences as in prior works in marketing. Provided that the predisposed thinking styles can be influenced by individuals' cultural backgrounds, the authors suggest cross-cultural studies as an approach to further validate the present findings.Practical implicationsGiven the recent trends that consumers are increasingly exposed to audiovisual ads with the rapid growth of various video-based platforms (e.g. YouTube) and mobile advertising, this empirical study may assist contemporary marketers in considering an acoustic strategy for brand communication using the audiovisual advertisement. This study suggests that the pitch of ad background music can serve as a manageable strategic tool that can assist in establishing an image of a competent brand.Originality/valueThis research highlights a seemingly-trivial element in audiovisual advertisements, the pitch of background music, as a crucial determinant of the perceived competence of an advertised brand upon which further brand evaluations (e.g. brand trust, purchase intention) are based. An important yet overlooked effect of ad recipients' predisposed thinking styles on how consumers respond to the changes of background cues in audiovisual commercials is also proposed.

Nowadays, commercial music has extreme loudness and heavily compressed dynamic range compared to the past. Yet, in music source separation, these characteristics have not been thoroughly considered, resulting in the domain mismatch between the laboratory and the real world. In this paper, we confirmed that this domain mismatch negatively affect the performance of the music source separation networks. To this end, we first created the out-of-domain evaluation datasets, musdb-L and XL, by mimicking the music mastering process. Then, we quantitatively verify that the performance of the state-of-the-art algorithms significantly deteriorated in our datasets. Lastly, we proposed LimitAug data augmentation method to reduce the domain mismatch, which utilizes an online limiter during the training data sampling process. We confirmed that it not only alleviates the performance degradation on our out-of-domain datasets, but also results in higher performance on in-domain data., Accepted to ISMIR 2022