At the heart of FIDO’s powerful leak detection solution is an AI algorithm that has been expertly tuned to recognise the auditory evidence of water leaks. With an accuracy rate of >92%, FIDO AI is the only technology that can decipher the size of leaks.
As a sensor agnostic, non-proprietary platform with a commitment to open data, FIDO doesn’t require third party acoustic material to recognise leaks. No matter how sophisticated your water network is – when it comes to speed and accuracy, FIDO is unrivalled.
Features of FIDO AI
FIDO uses its deep-learning algorithm to analyse a library of over 1.7 million audio samples to recognise leaks with >92% accuracy. With its ever-expanding collection of data, FIDO is constantly evolving – becoming more and more efficient at detecting leaks.
With no reliance on specific hardware and fully compliant with industry data sharing protocols – FIDO is safe, reliable and hundreds of times faster than humans at identifying leaks.
FIDO AI is continuously updated to accommodate new analytical features, developed in response to client needs.
Leak / no leak decisions to >92% accuracy: As FIDO collects and verifies files, accuracy improves.
Easy to use cloud-based platform: With an intuitive web-based dashboard, FIDO is platform user-friendly and highly secure.
Live and operational 24/7: If you ever have any queries, the system is supported by the FIDO app and their dedicated helpline.
Sensor agnostic: FIDO AI can process data from any source, including legacy sensors.
Leak sizing: As the world’s first technological solution that’s able to establish the size of a leak, FIDO is a true game changer for the water industry.
Unique leak library: As more clients use their AI, FIDO’s library of 1.7 million+ in-pipe and above-ground leak signatures grow rapidly.
Fully scalable with instant deployment: Pay for what you use with FIDO’s data-as-a-service (DaaS) model, then deploy it immediately.
Monthly subscription only: There are no upfront costs, only a monthly membership price per sensor on your network.