Publications

Extending the partial equivalent sound pressure level model to manage residual sound and exclude unwanted events identified with artificial intelligence

Long-term noise monitoring measurements are often affected by unwanted events, residual sound and other irrelevant sound sources, which can reduce the accuracy of the reported noise levels. This paper presents an extension of the partial equivalent sound pressure level model that combines corrections for residual sound and irrelevant sounds with automatic detection of unwanted events using artificial intelligence.

The proposed method is evaluated through a case study and compared with the procedures described in ISO 1996-2. The results indicate that the extended model can provide more representative estimates of equivalent sound pressure levels while improving the efficiency of data processing. The paper also discusses the assumptions, limitations and future development of the method.

Authors: Daniela Toledo Helboe, Karl Henrik Ejdfors and Trond Iver Pedersen, Norsonic AS

Evaluación del indicador nivel de presión sonora equivalente parcial en un escenario diferente de ruido ambiental

Accurately estimating equivalent sound pressure levels can be challenging when measurements are affected by irrelevant sounds. This study evaluates the Partial Equivalent Sound Pressure Level indicator in an environmental noise scenario that had not previously been investigated.

Results obtained from unattended noise monitoring stations are presented for relevant sound sources and compared across five different approaches for handling irrelevant sounds. The study highlights how the chosen method for managing irrelevant sounds can influence the final results and provides further insight into the applicability of the Partial Equivalent Sound Pressure Level indicator in environmental noise monitoring.

Mark: The paper is in Spanish.

Authors: Daniela Toledo Helboe, Jorge Paez Rodriguez and Trond Iver Pedersen, Norsonic AS

A comparative study of noise event identification using AI in unattended monitoring

This paper explores the use of multi-microphone devices and artificial intelligence (AI) for identifying noise events in unattended noise monitoring. The primary focus is to assess the reliability of a machine learning model initially trained on a dataset representing a particular soundscape. We evaluate the performance of this model when applied to diverse datasets collected from similar yet distinct soundscapes, encompassing various environmental conditions and noise profiles. Through comparative analysis, we determine the model’s adaptability and potential limitations. The findings of this study offer insights into how well AI-based noise event identification models can work in different situations. This lays the groundwork for enhancing their applicability in diverse real-world settings and improving how well unattended noise monitoring systems function.

Authors: Karl Henrik Ejdfors and Naru Sato, Norsonic AS, Lars Andreas Sæle, Norconsult Norge AS

Consideration and case study using a time-difference-of-arrival directional device for occupational noise measurement

To protect worker’s hearing in noisy workplaces, noise levels are measured periodically, and noise sources are located based on the results of recordings or video cameras. If the noise level is above a criterion level, noise abatement measures are taken, or workers are suggested to wear hearing protection. Measuring the noise level can be done simply by placing a sound level meter on the floor, but finding the noise abatement is a time-consuming task. It is necessary to listen to the recordings one by one or review the video to find the noise abatement. We believe that a simple method of estimating noise abatement is very useful, since some noisy workplaces do not allow recordings or video cameras for privacy and confidentiality reasons.

In this paper, noise source identification using a time-difference-of-arrival directional device in an indoor environment is examined. The results are analyzed in the context of room properties such as reverberation time, dimensions, and device localization. The paper also reports on practical usecases in actual sites.

Authors: Naru Sato, Karl Henrik Ejdfors and Erlend Fasting, Norsonic AS, Takeshi Nakaichi, RION Co., ltd. and Lars Andreas Sæle, Norconsult Norge

Partial equivalent sound pressure level as an approach to manage irrelevant sounds in environmental noise measurements

Residual sound and unwanted events are known factors affecting the accuracy of environmental noise measurements. ISO 1996-2 provides methods to manage these irrelevant sounds, for example by applying correction factors, but the methods require a degree of knowledge of the irrelevant sounds that is not always practical or possible to obtain when performing long-term measurements with unattended monitoring stations in complex urban soundscapes.

While automatic detection of irrelevant sounds provides the required information, it also allows for new approaches not described in ISO 1996-2. In this paper, we discuss a metric called partial equivalent sound pressure level (Partial Leq), calculated after data samples with residual sound and unwanted events are identified automatically and replaced by project-specific values. Our hypothesis is that better estimates of equivalent sound pressure levels (Leq) can be achieved with this metric, compared with the general methods given in ISO 1996-2. Comparisons of results are presented, together with a discussion of applicability of the Partial Leq metric and experiences gathered from its use.

Authors: Daniela Toledo and Trond Iver Pedersen, Norsonic AS

Automatic detection of source direction and exclusion of irrelevant sounds in unattended noise monitoring systems

A device featuring 8 MEMS microphones has been designed, which allows localizing dominant sound sources in 3D space by implementing techniques based on time difference of arrival. The device, also called Noise Compass, is intended to be used together with an outdoor measurement microphone in a noise monitoring terminal. By defining regions of interest in the vertical and horizontal planes in a monitoring and analysis system, non-relevant sounds can be automatically detected and excluded from the noise measurements. This article describes the direction detection mechanisms and shows two examples of system application: aircraft noise monitoring and a construction site with a road and railway nearby. Finally, other system applications are discussed.

Authors: Daniela Toledo and Erlend Fasting, Norsonic AS

AI-technology for efficient noise monitoring and analysis in complex urban soundscapes

NorCloud is a user-friendly software platform for noise monitoring and data analysis. Designed for use with advanced monitoring terminals that can localize dominant sound sources, it helps users manage projects involving multiple measurement locations and microphones.

The platform makes it easier to identify, classify and extract sound events from large datasets. AI-based processing can automatically remove non-relevant sound sources, reducing the time needed for data review and improving the efficiency of environmental noise assessments.

The article discusses the benefits and limitations of automated noise monitoring systems and explores current applications as well as future development opportunities.

Authors: Karl Henrik Ejdfors, Norsonic AS