 Hello everyone, I'm Marta Lennatti and in this video I'm going to briefly describe our project, introducing a novel hearing screening platform based on a speech and noise test and artificial intelligence. Speech and noise tests, by addressing speech recognition abilities in noise, are becoming increasingly popular, as hearing screening tests are able to promote awareness and early identification of hearing loss. These tests can be based on different kind of speech material and are able to offer a quick and self-administered tool for screening in remote modalities through mobile apps. Our speech and noise test uses meaningless vowel, consonant vowel, like ABBA or ATTA, allowing a reduced dependency on the native language. It is based on a free alternative force choice recognition task and it estimates speech recognition abilities in noise using an adaptive staircase procedure. To date, speech and noise tests are mainly based on the analysis of a single feature that is the speech reception threshold to predict the screening outcome. In addition to the SRT, estimated using a newly developed staircase, our system extracts a set of additional features. These features are fed into machine learning algorithms to implement a multivariate approach for automatic hearing loss detection. As a result, our multivariate classifier achieved better performance and sensitivity when compared to a conventional, univariate classifier based on SRT only. Ongoing research includes the validation of the procedure on a larger population, together with the implementation of an icon-based module to assess additional features. Feel free to contact us if you wish to collaborate.