ABSTRACT
Speech is a highly coordinated process that requires precise control over vocal tract morphology/motion to produce intelligible sounds while simultaneously generating unique exhaled flow patterns. The schlieren imaging technique visualizes airflows with subtle density variations. It is hypothesized that speech flows captured by schlieren, when analyzed using a hybrid of convolutional neural network (CNN) and long short-term memory (LSTM) network, can recognize alphabet pronunciations, thus facilitating automatic speech recognition and speech disorder therapy. This study evaluates the feasibility of using a CNN-based video classification network to differentiate speech flows corresponding to the first four alphabets: /A/, /B/, /C/, and /D/. A schlieren optical system was developed, and the speech flows of alphabet pronunciations were recorded for two participants at an acquisition rate of 60 frames per second. A total of 640 video clips, each lasting 1 s, were utilized to train and test a hybrid CNN-LSTM network. Acoustic analyses of the recorded sounds were conducted to understand the phonetic differences among the four alphabets. The hybrid CNN-LSTM network was trained separately on four datasets of varying sizes (i.e., 20, 30, 40, 50 videos per alphabet), all achieving over 95% accuracy in classifying videos of the same participant. However, the network's performance declined when tested on speech flows from a different participant, with accuracy dropping to around 44%, indicating significant inter-participant variability in alphabet pronunciation. Retraining the network with videos from both participants improved accuracy to 93% on the second participant. Analysis of misclassified videos indicated that factors such as low video quality and disproportional head size affected accuracy. These results highlight the potential of CNN-assisted speech recognition and speech therapy using articulation flows, although challenges remain in expanding the alphabet set and participant cohort.
ABSTRACT
Air pollution contains a complex mixture of poisonous compounds including particulate matter (PM) which has wide spectrum of adverse health effects. The main purpose of this study was to estimate the potential health impacts or benefits due to any changes in annual PM10 level in four major megacities of Iran. The required data of PM10 for AirQ software was collected from air quality monitoring stations in four megacities of Iran. The preprocessing was carried out using macro coding in excel environment. The relationship between different presumptive scenarios and health impacts was determined. We also assessed the health benefits of reducing PM10 to WHO Air Quality Guidelines (WHO-AQGs) and National Ambient Air Quality Standards (NAAQSs) levels with regard to the rate of mortality and morbidity in studied cities. We found that the 10 µg/m3 increase in annual PM10 concentration is responsible for seven (95% CI 6-8) cases increase in total number of deaths per 2 × 105 person. We also found that 10.7, 7.2, 5.7, and 5.3% of total death is attributable to short-term exposure to air pollution for Ahvaz, Isfahan, Shiraz, and Tehran, respectively. We found that by attaining the WHO's proposed value for PM10, the potential health benefits of 89, 84, 79, and 78% were obtained in Ahvaz, Isfahan, Shiraz, and Tehran, respectively. The results also indicated that 27, 10, 3, and 1% of health impacts were attributed to dust storm days for Ahvaz, Isfahan, Shiraz, and Tehran, respectively.