Interspeech

19,244 papers5,917 directions
Interspeech is a research venue dedicated to the science and technology of spoken language processing. Within the machine learning and artificial intelligence community, it serves as a forum for advancing techniques in audio signal analysis, speech recognition, and synthesis. The venue's contributions often focus on the intersection of deep learning architectures and acoustic modeling, emphasizing the development of systems that can interpret and generate human speech across diverse environments. Research published at Interspeech frequently explores end-to-end learning from raw audio waveforms and the integration of local and global contexts within speech architectures. Technical developments include the use of asymmetric convolutions for time-frequency processing and multi-resolution architectures for sound event detection. Recent work also addresses robust speech processing in out-of-distribution environments and the application of self-supervised learning to enhance user privacy in automatic speech recognition. The venue covers a broad range of applications, including speech enhancement through knowledge distillation, voice conversion using contrastive predictive coding, and speaker diarization. It also features research on specialized tasks such as the automatic detection of speech sound disorders in children and mitigating household confusability in speaker identification systems. Prolific contributors to the venue include researchers such as Chi-Chun Lee, Yu Tsao, and Hsin-Min Wang, who represent the ongoing effort to refine how artificial intelligence models process complex auditory data.

Papers by Year

2026 5
2025 1,156
2024 1,036
2023 1,110
2022 1,122
2021 979
2020 1,014
2019 948
2018 774
2017 825
2016 800
2015 782
2014 630
2013 792
2012 675
2011 845
2010 772
2009 760
2008 756
2007 747
2006 656
2005 860
2004 766
2003 125
2002 107
2001 83
2000 132

Top Research Directions

Research directions most represented by papers from this venue.

The Single Speaker Generalization Bottleneck in Speech Synthesis 522
Generalizing from Limited Cohorts in Speech Research 439
Robust Natural Language Understanding for Noisy Speech Transcripts 399
Mitigating Performance Degradation in Noisy Speech Recognition 389
Generalization to Unseen Acoustic Environments 366
Overcoming the Stationary Noise Assumption in Signal Enhancement 353
Partitioning Variance with Linear Mixed Effects Models 337
Multi-Domain Feature Fusion for Robust Speech Analysis 329
Overcoming Data Scarcity in Low-Resource Speech Recognition 290
Personalizing Speech Models with Minimal Data 280
Speaker Adaptation for Personalized Speech Recognition 259
Synchronized Acoustic and Articulatory Speech Modeling 256
Bridging the Gap Between Read and Spontaneous Speech 252
Temporal Alignment of Speech and Text 241
Open Vocabulary Recognition for Spoken Language Systems 230

Top Authors

AuthorPapers
Shrikanth Narayanan237
Haizhou Li212
Shinji Watanabe166
Helen M. Meng99
Dong Yu90
EngSiong Chng80
Lin-shan Lee80
Jianhua Tao77
Hung-yi Lee76
Yanmin Qian76
Yonghong Yan76
Najim Dehak73
Jinyu Li70
Florian Metze65
Kong Aik Lee59
Zhen-Hua Ling59
Dilek Hakkani Tur56
James R. Glass56
Zhiyong Wu56
Jun Du52

Impactful Papers

Papers from this venue that shaped research clusters.

Foundational

Influential

Related Venues

Venues with the most similar research cluster distributions.

Papers

TitleYear
Automatic language identification using long short-term memory recurrent neural networks2014
Automatic modelling of depressed speech: relevant features and relevance of gender2014
Automatic recognition of attitudes in video blogs - prosodic and visual feature analysis2014
Automatic recognition of speaker physical load using posterior probability based features from acoustic and phonetic tokens2014
Automatic speech feature classification for children with cochlear implants2014
Automatic speech recognition and translation of a Swiss German dialect: Walliserdeutsch2014
Automating an objective measure of pediatric speech intelligibility2014
Autoregressive product of multi-frame predictions can improve the accuracy of hybrid models2014
A whispered Mandarin corpus for speech technology applications2014
Backoff inspired features for maximum entropy language models2014
Bayesian calibration for forensic evidence reporting2014
Bayesian factorization and selection for speech and music separation2014
Beyond cross-entropy: towards better frame-level objective functions for deep neural network training in automatic speech recognition2014
Binary mask estimation based on frequency modulations2014
Binaural deep neural network classification for reverberant speech segregation2014
BioKIT — real-time decoder for biosignal processing2014
Blind source extraction based on a direction-dependent a-priori SNR2014
Blind speech source localization, counting and separation for 2-channel convolutive mixtures in a reverberant environment2014
Boosted deep neural networks and multi-resolution cochleagram features for voice activity detection2014
Boosting bonsai trees for efficient features combination: application to speaker role identification2014
Boundary contraction training for acoustic models based on discrete deep neural networks2014
Building a naturalistic emotional speech corpus by retrieving expressive behaviors from existing speech corpora2014
Building a vocabulary self-learning speech recognition system2014
Building resources for Algerian Arabic dialects2014
BUT 2014 Babel system: analysis of adaptation in NN based systems2014
Can adolescents with autism perceive emotional prosody?2014
Canonical correlation analysis and local fisher discriminant analysis based multi-view acoustic feature reduction for physical load prediction2014
Chaotic mixed excitation source for speech synthesis2014
Choosing useful word alternates for automatic speech recognition correction interfaces2014
Classification of cognitive load from speech using an i-vector framework2014
Cluster based Chinese abbreviation modeling2014
Co-channel speech detection via spectral analysis of frequency modulated sub-bands2014
Collecting a corpus of Dutch noise-induced `slips of the ear'2014
Combination of FST and CN search in spoken term detection2014
Combination of multilingual and semi-supervised training for under-resourced languages2014
Combining recurrent neural networks and factored language models during decoding of code-Switching speech2014
Combining source and system information for limited data speaker verification2014
Combining tandem and hybrid systems for improved speech recognition and keyword spotting on low resource languages2014
Comparing approaches to convert recurrent neural networks into backoff language models for efficient decoding2014
Comparing decoding strategies for subword-based keyword spotting in low-resourced languages2014
Comparing parameterizations of pitch register and its discontinuities at prosodic boundaries for Hungarian2014
Comparing reaction time sequences from human participants and computational models2014
Comparing time-frequency representations for directional derivative features2014
Comparison of speech quality with and without sensors in electromagnetic articulograph AG 501 recording2014
Comparison of vocal tract transfer functions calculated using one-dimensional and three-dimensional acoustic simulation methods2014
Component structuring and trajectory modeling for speech recognition2014
Concept-to-speech generation by integrating syntagmatic features into HMM-based speech synthesis2014
Consonant context effects on vowel sensorimotor adaptation2014
Constrained speaker linking2014
Content matching for short duration speaker recognition2014

« prev Page 214/385 next »