Output details

Automatic Detection and Recognition of Tonal Bird Sounds in Noisy Environments

This paper introduces a new approach in an emerging field of computer analysis of bird acoustic data. This, for the first time, enables automatic detection of bird tonal vocalisations in noisy environments and provides feature representation that shows significant performance advantages over the current state-of-the-art. This work was part funded under an INDEX grant, provided by Advantage West Midlands and EPSRC, in collaboration with Biocensus Ltd. It was the subject of an article in ‘The Engineer’. This work has led to collaboration with Ohio State University (USA) and to commercial exploitation discussions with Biocensus (UK), Agant (UK), Rydell Consulting (Sweden).