Noise pollution is one of the topmost quality-of-life issues for urban residents in Ireland. While the World Health Organisations’ safety threshold for sound is 40dB, noise in Dublin exceeds 55 decibels at night and surpasses 70 decibels during the day. With piling evidence from medical studies about the detrimental impact noise pollution is having on our health (diabetes, hearing loss, cardiovascular disease), noise pollution is no longer something we can sound out.
Gemmo AI has launched a new AI tool that will change how we monitor noise. The tool is designed to empower noise sensors with AI technology, enabling them to generate valuable insights through data analytics SaaS. At the moment, in areas with high noise concentrations, noise is measured by sensors and analysed manually.- This is massively time-consuming, taking hours to process minutes of audio and is error-prone.
Gemmo says this is part of the reason we are falling behind in terms of reducing noise pollution.
According to CEO Luca Marchesotti “Speed is a huge issue for the environment. The longer we take to address important issues, the bigger they get and the less attention they receive. These kinds of issues call for smart innovation. That’s why we’ve developed AI to speed up Noise and Vibration assessment. It works by automatically finding and labelling relevant events, and discarding background signals”
At the heart of Gemmo AI’s philosophy is the belief that every sensor produces data, but most of them are unable to offer a comprehensive and in-depth analysis of that data. Manual work and workforce effort are required to extract useful information from the data generated by these sensors, involving several stages along the value chain from the sensor manufacturers upstream to the industries downstream that require analytics on acoustics.
“This manual effort and investment represent the kind of practical issues that are putting a halt to change and roadblocking how we challenge noise pollution,” continued Luca.
Gemmos sound algorithms offer several capabilities that make them an essential tool for noise monitoring. It provides noise source attribution, which quickly discards irrelevant noise clips and detects acoustic events such as blasts, rain, and wind. This makes it possible to remove background noise, such as rain and wind, from large audio clips automatically, and identify sparse, rare sound events of interest, such as blasting, piling, and engine spinning.
Earlier this year, Gemmo AI launched with the goal of injecting AI into the environmental monitoring industry, at a large scale. Now, with the release of their Sound Analysis APIs, they’re making serious inroads in achieving that goal, helping the fight against noise pollution.
Marchesotti hopes Gemmo AI will lead the way in this industry, adding:
“We want to be the leading AI company doing this. Industries are shifting their views on AI and this is just one case where we can build AI for good. In tackling things like noise pollution, we’re showing people that AI isn’t this evil thing we should fear, but a really effective and useful tool in building a better world”
Gemmo have already started trialling their sound analysis APIs with incredible results.
See more stories here.