ICASSP

17,183 papers3,342 directions
ICASSP is a research venue focused on the intersection of signal processing and machine learning. Its publications cover a broad range of technical domains, including speech recognition, audio source separation, and computer vision. The venue serves as a platform for research into the processing of acoustic, visual, and temporal data through computational models. Recent contributions include work on speaker-aware training, spatial filtering for speech enhancement, and the automated scaling of video-language datasets. Within the machine learning community, ICASSP is known for its emphasis on feature engineering, model optimization, and robust pattern recognition. Research directions represented at the venue include geometric feature packing, neural collapse, and maximum entropy regularization for out-of-distribution detection. The venue also features studies on few-shot learning, task-specific data compression for machine vision, and the application of transformers for open-set embedding adaptation in speaker identification. The venue’s output includes practical applications such as Alzheimer’s disease diagnosis via speech analysis, stock portfolio selection using hypergraph-based reinforcement learning, and defect visual inspection through self-supervised learning. Prolific contributors to this research space include Hirokazu Kameoka, Shogo Seki, Ian McLoughlin, and Li-Rong Dai. These works often address technical trade-offs, such as managing the balance between signal suppression and distortion in audio processing or improving semantic segmentation in non-rigid scenes using deformable convolutional neural networks.

Papers by Year

2026 13
2025 1,672
2024 1,557
2023 1,656
2022 1,183
2021 1,056
2020 1,051
2019 870
2018 703
2017 593
2016 554
2015 459
2014 595
2013 613
2012 483
2011 442
2010 443
2009 397
2008 357
2007 372
2006 325
2005 329
2004 318
2003 284
2002 173
2001 144
2000 142
1999 100
1998 98
1997 112
1996 82
1995 86
1994 65
1993 80
1992 53
1991 67
1990 43
1989 34
1988 46
1987 28
1986 18
1985 15
1984 13
1983 12
1982 6
1980 1

Top Research Directions

Research directions most represented by papers from this venue.

Learning High-Level Acoustic Structures from Unlabeled Audio 27
The Perceptual Evaluation Gap in Generative Audio 26
Generalization to Unseen Acoustic Environments 24
Cross-Task Generalization in Audio and Speech Processing 22
Neural Speech Processing with Variable Speaker Cardinality 22
Resolving Overlapping Speech in Multi-Speaker Audio 22
Joint Optimization of Speech Enhancement and Recognition 21
Visual Anchoring for Robust Speech Enhancement 20
Robust Audio Modeling Under Acoustic Domain Shift 20
Factorizing Speech into Content Timbre and Prosody 19
The Acoustic Simulation to Reality Gap 17
Hybrid Time-Frequency Loss Optimization 16
Semidefinite Relaxation for Non-Convex Engineering Optimization 15
Balancing Noise Suppression and Speech Distortion 15
Audio Source Separation Without Clean Ground Truth 15

Top Authors

AuthorPapers
Georgios B. Giannakis166
Shinji Watanabe164
Alfred O. Hero162
Haizhou Li153
Shrikanth Narayanan133
Dong Yu105
Helen M. Meng94
Phil Woodland88
Rama Chellappa86
Jinyu Li78
Lin-shan Lee76
Wing-Kin Ma76
Yanmin Qian76
Kai Yu74
Zhiyong Wu74
Wenwu Wang69
EngSiong Chng68
Miki Haseyama66
Hung-yi Lee65
Ananthram Swami62

Impactful Papers

Papers from this venue that shaped research clusters.

Foundational

Influential

Related Venues

Venues with the most similar research cluster distributions.

Papers

TitleYear
The waveform segment vocoder: A new approach for very-low-rate speech coding1985
Alternatives to the generalized cross correlater for time delay estimation1984
A model based approach for 2-D MEPS analysis1984
An iterative algorithm for robust 2-D spectrum estimation1984
An iterative method for restoring noisy blurred images1984
Classification of textures using Markov random field models1984
Improved hidden Markov modeling of phonemes for continuous speech recognition1984
Investigation of text-independent speaker identification techniques under conditions of variable data1984
Morphological skeleton representation and coding of binary images1984
Multichannel linear predictive coding of color images1984
On hardware description from block diagrams1984
Speaker normalization algorithms for very-low-rate speech coding1984
Synthesis by rule of english intonation patterns1984
Two special digital processing systems for recognition of moving objects1984
A Chinese speech recognition system1983
A comparison of methods for 300-400 b/s vocoders1983
A multiple model algorithm for the adaptive restoration of images1983
A segment vocoder at 150 b/s1983
Further investigation of probabilistic methods for text-independent speaker identification1983
Image restoration using order-constrained least-squares methods1983
Iterative deconvolution using several different distorted versions of an unknown signal1983
Maximum likelihood estimation of composite source models for image coding1983
The converging squares algorithm: An efficient multidimensional peak picking method1983
The wavelet reconstructed with phase-only and a study of its uniqueness1983
Two-dimensional linear predictive analysis of arbitrarily-shaped regions1983
Two-dimensional spectral estimation using spatial autoregressive models1983
A variable-order Markov chain for coding of speech spectra1982
Composite source coding techniques for image bandwidth compression1982
Nonlinear filtering using linear combinations of order statistics1982
Segment quantization for very-low-rate speech coding1982
Some experiments in ADPCM coding of images1982
The application of probability density estimation to text-independent speaker identification1982
A 4800 bps LPC vocoder with improved excitation1980

« prev Page 344/344