A working definition of Computational Audio (CA) is the intersection of Computer Science with (digital) audio analysis, processing, and synthesis by Digital Signal Processing. Because Computer Science reaches into the Humanities and Social Sciences, CA also encompasses aspects of that disciplinary territory. CA is concerned with the transition from audio data to audio information, and to knowledge and understanding for all forms of audio, i.e., speech, music and environmental. [NB: we do however exclude computer speech technology which is well-established and comprehensively covered in many forums.] Computational Audio seeks to make this information and knowledge useable in various application domains, including creative activity. Thus it is also concerned with the inverse process, from information and knowledge to acoustic transmission for human perception, enlightenment, and entertainment.
Relevant aspects of Computer Science include knowledge representation (including ontologies), inference, information retrieval, machine learning, artificial intelligence, cryptography, autonomous agents, Human Computer Interaction, cognition & perception, communication theory, network delivery, and more.
Relevant aspects of Audio include musical, physical & architectural acoustics, music performance, sound engineering, signal representation, and signal processing.
Relevant aspects of the humanities and social sciences include: musicology, music theory, anthropology, ethnography, psychology, philosophy, cognition and perception, and more.
Computational Audio is an appropriate topic because it appeals on many levels: to us as individuals as it deals largely with a creative art (music) that is prevalent across all societies; to scientists and engineers because it incorporates many aspects of computational science and informatics; it is representative of born-digital content, remaining digital end to end; it supports investigation into intelligent information architectures via advanced search and/or navigation requirements. Music and audio is thus a valuable vector for presenting advanced informatics to a wide audience.
While several topics (like immersive audio, physical modelling of instruments and audio signal processing) are well established, they have grown in recent years from laboratory curiosities into important and fast-changing topics, particularly as Augmented and Virtual Reality gains strength.
Alongside this there are several new aspects to the field. This includes tangible interfaces to digital musical instruments (extending to wearable interfaces). The burgeoning use of Semantic Web technologies, both to represent musical and acoustic knowledge and to construct global-scale, de-centralised intelligent information systems has tremendous potential to disrupt the market in recorded music, whilst being a vibrant research topic. Environmental Audio is a new field that encapsulates topics including acoustics, signal processing and machine learning, and offers great promise in helping society deal many different acoustic situations, from intruder alerts to biodiversity.
A working definition of Computational Audio (CA) is the intersection of Computer Science with (digital) audio analysis, processing, and synthesis by Digital Signal Processing. Because Computer Science reaches into the Humanities and Social Sciences, CA also encompasses aspects of that disciplinary territory. CA is concerned with the transition from audio data to audio information, and to knowledge and understanding for all forms of audio, i.e., speech, music and environmental. [NB: we do however exclude computer speech technology which is well-established and comprehensively covered in many forums.] Computational Audio seeks to make this information and knowledge useable in various application domains, including creative activity. Thus it is also concerned with the inverse process, from information and knowledge to acoustic transmission for human perception, enlightenment, and entertainment.
Relevant aspects of Computer Science include knowledge representation (including ontologies), inference, information retrieval, machine learning, artificial intelligence, cryptography, autonomous agents, Human Computer Interaction, cognition & perception, communication theory, network delivery, and more.
Relevant aspects of Audio include musical, physical & architectural acoustics, music performance, sound engineering, signal representation, and signal processing.
Relevant aspects of the humanities and social sciences include: musicology, music theory, anthropology, ethnography, psychology, philosophy, cognition and perception, and more.
Computational Audio is an appropriate topic because it appeals on many levels: to us as individuals as it deals largely with a creative art (music) that is prevalent across all societies; to scientists and engineers because it incorporates many aspects of computational science and informatics; it is representative of born-digital content, remaining digital end to end; it supports investigation into intelligent information architectures via advanced search and/or navigation requirements. Music and audio is thus a valuable vector for presenting advanced informatics to a wide audience.
While several topics (like immersive audio, physical modelling of instruments and audio signal processing) are well established, they have grown in recent years from laboratory curiosities into important and fast-changing topics, particularly as Augmented and Virtual Reality gains strength.
Alongside this there are several new aspects to the field. This includes tangible interfaces to digital musical instruments (extending to wearable interfaces). The burgeoning use of Semantic Web technologies, both to represent musical and acoustic knowledge and to construct global-scale, de-centralised intelligent information systems has tremendous potential to disrupt the market in recorded music, whilst being a vibrant research topic. Environmental Audio is a new field that encapsulates topics including acoustics, signal processing and machine learning, and offers great promise in helping society deal many different acoustic situations, from intruder alerts to biodiversity.
