International Conference on Acoustics, Speech, and Signal Processing

International Conference on Acoustics, Speech, and Signal Processing
Name	International Conference on Acoustics, Speech, and Signal Processing
Acronym	ICASSP
Discipline	Signal processing
Publisher	Institute of Electrical and Electronics Engineers
First	1976
Frequency	Annual

International Conference on Acoustics, Speech, and Signal Processing is an annual flagship forum for research in Institute of Electrical and Electronics Engineers, IEEE Signal Processing Society, speech recognition, image processing, wireless communications, machine learning. The conference gathers researchers from Bell Labs, Massachusetts Institute of Technology, Stanford University, University of California, Berkeley, University of Cambridge and industry labs such as Google, Microsoft Research, IBM Research, Facebook AI Research to present advances in Fourier transform, Kalman filter, hidden Markov model, deep neural network methods. ICASSP has shaped technical directions involving contributors affiliated with National Institute of Standards and Technology, European Research Council, National Science Foundation, Chinese Academy of Sciences, Tsinghua University.

History

The meeting series originated in the 1970s when engineers from Bell Labs, AT&T, Hughes Aircraft Company, Texas Instruments and universities such as University of Illinois Urbana-Champaign, Princeton University, Carnegie Mellon University coordinated workshops on acoustics, speech processing and signal processing; early conferences featured keynote speakers from Nokia and Ericsson and exhibitors like RCA and Philips. Over decades ICASSP featured seminal demonstrations from researchers affiliated with Massachusetts Institute of Technology, Stanford University, University of Oxford, California Institute of Technology, University of Tokyo and fostered collaborations with projects at Los Alamos National Laboratory, Lawrence Berkeley National Laboratory, Sandia National Laboratories.

Scope and Topics

The conference covers algorithmic and theoretical work from communities including speech recognition, audio engineering, biomedical signal processing, remote sensing, radar, sonar, communications, computer vision, natural language processing, pattern recognition. Typical tracks draw submissions using methods rooted in probability theory from researchers at Columbia University, Yale University, University of Michigan, Johns Hopkins University, Cornell University and applications tied to NASA, European Space Agency, Siemens, Bosch, Qualcomm.

Organization and Sponsorship

ICASSP is organized under the aegis of the IEEE Signal Processing Society with local arrangements by host institutions such as University of Toronto, McGill University, Imperial College London, University of Sydney, Seoul National University. Sponsorship regularly involves corporations including Intel Corporation, NVIDIA, Amazon, Apple Inc. and funding or endorsement from agencies such as the National Institutes of Health, Defense Advanced Research Projects Agency, European Commission, Australian Research Council.

Proceedings and Publications

Accepted papers are published in the IEEE conference proceedings and indexed by databases that serve ACM Digital Library, IEEE Xplore, Scopus, Web of Science; extended versions often appear in IEEE Transactions on Signal Processing, IEEE Transactions on Audio, Speech, and Language Processing, Signal Processing, Journal of the Acoustical Society of America. Special issues have been coordinated with editorial offices at Elsevier, Springer, Oxford University Press and repositories such as arXiv host preprints by authors from University of Washington, University of Southern California, Peking University.

Notable Papers and Contributions

Landmark contributions presented at ICASSP include early formulations of the hidden Markov model for speech recognition by teams from IBM Research and Bell Labs, advances in cepstral analysis by researchers at University of California, Los Angeles and Royal Institute of Technology (KTH), innovations in beamforming and adaptive filtering from Georgia Institute of Technology and University of Texas at Austin, demonstrations of deep learning architectures by groups affiliated with University of Toronto, Carnegie Mellon University, Google DeepMind that influenced later publications in Nature and Science. Other influential works involved compressive sensing with contributors from Rice University, Yale University, Princeton University and source separation techniques by teams at Imperial College London and École Polytechnique Fédérale de Lausanne.

Awards and Special Sessions

The conference regularly features awards and recognitions coordinated with the IEEE Signal Processing Society such as best paper awards, best student paper awards, and plenary lectures delivered by fellows of IEEE, Royal Society, National Academy of Engineering, National Academy of Sciences. Special sessions have included invited panels on topics tied to European Research Council initiatives, workshops co-located with meetings of International Telecommunication Union, ICASSP-hosted challenges in collaboration with industry partners like Google, Amazon Web Services, NVIDIA and competitions similar to those run by ImageNet and Kaggle.

Conferences by Year and Location

Past ICASSP venues have included major cities and institutions: 1980s meetings at Princeton University, University of Southern California, University of Illinois Urbana-Champaign; 1990s sites such as San Francisco, Kyoto, Munich; 2000s editions in Istanbul, Philadelphia, Honolulu, Las Vegas; 2010s conferences hosted by Vancouver, Florence, Shanghai, Brisbane; recent editions have taken place in Toronto, Barcelona, Singapore, Athens with organizational involvement from University College London, National University of Singapore, University of Patras, Hong Kong University of Science and Technology.

Category:Academic conferences