A sound recognition method with incrementally taking sound stream into consideration
Mamoru Iwaki, Shun Nakayama · 2016
Domestic sound recognition is an important ability for our daily lives. Although we can recognize sounds as they arrive, conventional automatic sound/speech recognition systems cannot behave in a similar manner. This paper proposes a domestic sound recognition method that can incrementally take sound streams into consideration during the progress of time. This is useful for advancing human-machine interfaces, especially for hearing disabled persons.