A detailed look at speech recognition in realistic listening scenarios
In clinical audiology we are interested in optimal participation in social interactions of listeners with impaired hearing, hence we strive for optimizing auditory function. Speech recognition is a major component auditory function in the context of societal participation. A lot of research has been done in the field of speech recognition in realistic environments. However, in those studies the definition of “realistic” seems to be constructed based on theoretical considerations.
We are investigating and analysing bilateral recordings in realistic environments, e.g., at home, in public transport, in a shop, in a restaurant, outside . We will report on the nature of those recordings, in particular: 1) how “speech-like” are these recordings, and 2) how are binaural parameters distributed in these recordings.