A compelling use-case for Edge AI is the ability to perform real time classification of audio on low-power devices. By enabling processing of audio data on an edge device, latency, power-usage and cost can be reduced while maintaining a strong protection of end-users’ privacy. Even better, this functionality can be used to increase car and road safety.
Slovakia-based Inovasense have, during the course of a three month project, evaluated Imagimob AI as a platform for building audio classification applications targeting the automotive industry. Inovasense is an embedded hardware and software design house, specializing in the development of ultra low power devices with built-in hardware security.
Within the project, Inovasense engineers built, trained and evaluated several advanced AI models using Imagimob AI. Inovasense were quickly able to import their pre-existing datasets into the platform, visualize the data to find flaws in it, and then use Imagimob AI to evaluate new methods of data processing.
Regarding the data visualization module of the platform, Timotej, a senior software engineer at Inovasense says: “I really love this! The visualization capabilities of Imagimob AI were above my expectations.”
Once the data was in good shape the focus shifted to model-building and -training. In this part of the project several advanced AI models were built and evaluated inside Imagimob AI with good results. Asking Timotej about this part of the service he says: “I really like this, now I have much more information on what is going on with my models than I had before”.
In the next project phase Inovasense will use Imagimob AI to package the most promising AI models for testing on final hardware. Vladimir Vician, CEO at Inovasense says: “We are now up-to-speed using Imagimob AI and can use it to run Edge AI-projects for our customers. We are also considering using it in our own product development.”