Activities of the Vehicle Exterior Noise Division Committee of the Society of Automotive Engineers of Japan

Toru Yamazaki (Kanagawa University)

The Exterior Vehicle Noise Division Committee was established in FY2021 to address various issues related to vehicle noise. Its goals include developing methods to predict and evaluate noise as a source of disturbance, assessing its environmental impact, and understanding its effects on nearby residents’ awareness and reactions. This presentation highlights the committee's progress since its inception.

Subjective evaluation of a single motor vehicle noise with low frequency components

kazuma Yabuuchi (Kanagawa University Graduate School)･Shigenori Yokoshima (Kanagawa Environmental Research Center/Kanagawa University)･Makoto Morinaga (Daido University)･Koichi Makino･Tetsuya Doi･Sakae Yokoyama･Tomohiro Kobayashi (Kobayasi Institute of Physical Research)･Toru Yamazaki (Kanagawa University)

The authors conducted a subjective evaluation experiment in a low-frequency sound chamber. Single pass-by noises emitted by heavy vehicles with a predominant sound pressure level in the low-frequency range of 40 Hz or 50 Hz were used as stimuli. In this paper, we discussed the influence of the sound pressure level of the predominant frequency in the low-frequency component on the oppressive or vibratory feeling, which is characterized as the effect of low-frequency sound, and on the discomfort feeling due to the noise.

Objective evaluation of awaking due to road traffic noise
-Basic study using wearable devices-

Makoto Morinaga (Daido University)･Shigenori Yokoshima (Kanagawa Environmental Research Center / Kanagawa University)･Yoshiki Umezaki (Creative Research and Planning)･Toru Yamazaki (Kanagawa University)

Traditional studies on the effects of traffic noise on sleep have often used subjective surveys, but there's a demand for more objective scientific approaches. This research explores a simple and objective method to assess sleep disturbances by using wearable devices to detect awakenings and analyze their correlation with noise levels.

Study on Traffic Noise Data Generation Using Generative AI
-Possibility of generating traffic noise by manipulating parameters in latent space-

ManYong Jeong (National Institute of Technology, Numazu College)･Ritsuki Matsunaga (Advanced Course, National Institute of Technology, Numazu College)

This study examines the feasibility of applying generative AI technologies—such as Generative Adversarial Networks (GANs) and diffusion models—to flexibly and accurately generate traffic noise data that are difficult to record or obtain in real-world environments. The demand for large-scale, diverse datasets to support the development of traffic noise prediction models and the evaluation of noise abatement measures has been growing. However, direct collection of such data is often hampered by high measurement costs, environmental constraints, and privacy or security concerns.
In this research, we use existing measured data as training input to synthesize various traffic noise waveforms using generative AI. More specifically, we construct a conditional generative model that integrates meta-information—including traffic volume, vehicle mix, running speed, and weather conditions—to produce realistic noise waveforms. We then evaluate whether these generated signals can faithfully replicate key acoustic properties such as spectral and temporal characteristics. Furthermore, we verify the utility of the proposed method by applying the generated data to the training of noise prediction models and the evaluation of anomaly detection algorithms.
Ultimately, this study aims to demonstrate that improving the quality of simulated data via generative AI technologies can enhance the foundational capabilities for traffic noise analysis and inform the development of related policies.

Outline of road traffic noise prediction model “ASJ RTN-Model 2023”

Yasuaki Okada (Meijo University)･Katsuya Yamauchi (Kyushu University)･Shinichi Sakamoto (The University of Tokyo)

The Acoustical Society of Japan (ASJ) has had the Research Committee on Road Traffic Noise to develop a prediction model for road traffic noise for more than 50 years. As a result of the research activities, a new version of model, “ASJ RTN-Model 2023”, was published in last April. It is an upgrade version of the previous model, “ASJ RTN-Model 2018”, proposed in 2019. In developing the latest version, existing knowledge was widely taken into account, in particular, the sound power levels of road vehicles under non-steady running conditions and calculation methods for sound propagation were improved in wide range.