The following paper presents a novel method for the exploitation of urban noise and sound measurement. The overarching objective of this work is to advantageously mine & exploit the information embedded into urban sounds. Effectively contributing to the development of services and products via the advancement of the State of The Art in technology of information and telecommunication.
Traditionally, the grand majority of city and urban noise/sound is measure, analysed and classified with the purpose of drawing appropriate government legislation and regulations aimed at contributing to a healthier environment for humans. Furthermore, noise measurements are often, if not always, measured and processed with dedicated high-end microphones and devices. Costly microphones and processing hardware, often based on limited data samples, limit the extension of the exploitation of the data campaigns. Effectively, the trading of sound-related information is almost often a simple absolute noise level measurement business between a sound expert and a council-like or government-like institution.
With these premises, we can safely say that most of the urban noise and sound business is carried out with the sole purpose of reporting or denouncing, to the appropriate authorities, a misconduct (e.g. noisy streets, bars, cities) or correct a misuse of council resources (e.g. bus re-routing, city planning). Other types of urban noise business activities clearly exist but its span is definitely limited.
We believe that urban sounds do carry more information than what it is extracted to date. The wide availability of powerful AI tools makes it even more true. We believe that urban sound data can be conveniently mined via modern sound algorithm methods. We also believe that in the traditional way of processing urban sound datasets, the employment of costly hardware does represent a limiting factor for the business that can be built around urban sound analysis.
We present a technologically novel method for the capturing, processing and trading of urban sound-based information. The presented method is developed around consumer-grade sound devices that, being relatively inexpensive, can be considered a commodity hardware. An example that we would like to present as such is Echo, the cloud-based commercial multi-microphone smart speaker product by Amazon. Being widely available and easily purchased for US$60 over the Internet, Echo represents a great example of a commodity hardware that can be purchased or rented for virtually no cost.
Unlike traditional urban sound processing systems, we discuss how to move the sound processing algorithms to a cloud-based server. In a way, similarly to what Amazon did with the cloud product Alexa, we propose to decentralize part of the software processing of sound/noise to a more agile cloud-based server. In the following way we will be able to enable the use of these algorithms to a third party via the use of multiple APIs. This will naturally free the consumer-grade sound sensors from the difficult task of processing, making it extremely inexpensive.
In this paper, beside primarily focusing on the scientific aspect of the method, we will like to discuss as well, perhaps in greater details than what it is normally done, the main business scenarios where the presented idea could enable novel and exciting commercial opportunities. Hence, we will show that such a method will open new business models in the area of urban sound analysis and in its exploitation.
The novelty aspects of the approach presented in this proposal is manifold:
Instead of using proprietary and costly devices, we propose to explore the use of a consumer-grade commodity hardware for sound capturing and measurement.
Develop and implement de-centralized urban sound analysis algorithms for the processing of urban sound samples that are cloud-based. As such, their use will be offered via an edge server and will not depend, or will only marginally depend, on the capabilities of the employed sound capturing device. Effectively making it simpler and cheaper.
Build an overall system that is heavily cloud-based (e.g. edge computing)so that the information collected by the inexpensive sound capturing devices can be processed by professional remote sound algorithms, developed by independent experts, via APIs that are accessed following a pay-per-use business scheme.
From the business development point of view, the proposed overall system design structure is the most innovative aspect of this paper. Furthermore, the fact that all processing algorithms are basically cloud-based services, allows us to envision a large number of business cases effectively being enabled by what we propose. A detailed description of such an important aspect is given in further sections.