New MAP estimators for speaker recognition

We report the results of some experiments which demonstrate that eigenvoice MAP and eigenphone MAP are at least as effective as classical MAP for discriminative speaker modeling on SWITCHBOARD data. We show how eigenvoice MAP can be modified to yield a new model-based channel compensation technique which we call eigenchannel MAP. When compared with multi-channel training, eigenchannel MAP was found to reduce speaker identification errors by 50%.

• Publication year

  2003

• Venue

  Interspeech

• Publication date

  2003-09-01

• Fields of study

  Computer Science

• Identifiers
  DOI 10.21437/Eurospeech.2003-759
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Structural Gaussian mixture models for efficient text-independent speaker verification

  2002cited by this paper
• HIERARCHCIAL GAUSSIAN MIXTURE MODEL FOR SPEAKER VERIFICATION

  2002cited by this paper
• Maximum likelihood estimation of eigenvoices and residual variances for large vocabulary speech recognition tasks

  2002influential reference
• Speaker verification with data fusion and model adaptation

  2002cited by this paper
• A structural Bayes approach to speaker adaptation

  2001cited by this paper
• A Comparison of Adaptation Methods for Speaker Verification

  2001cited by this paper
• Speaker veri ® cation over the telephone q

  2000cited by this paper
• Bayesian adaptation revisited

  2000cited by this paper
• The NIST speaker recognition evaluation - Overview, methodology, systems, results, perspective

  2000cited by this paper
• Speaker verification over the telephone

  2000cited by this paper
• Speaker Verification Using Adapted Gaussian Mixture Models

  2000cited by this paper
• Unsupervised On-Line Adaptation in Speaker Verification

  2000cited by this paper
• Speaker identification and verification using eigenvoices

  2000cited by this paper
• Object Matching Using Deformable Templates

  1996cited by this paper
• Experiments with speaker verification over the telephone

  1995cited by this paper
• Maximum a posteriori estimation for multivariate Gaussian mixture observations of Markov chains

  1994cited by this paper

• PhiNet: Speaker Verification With Phonetic Interpretability

  2026cites this paper
• Overview of Speaker Modeling and Its Applications: From the Lens of Deep Speaker Representation Learning

  2024cites this paper
• 13 years of speaker recognition research at BUT, with longitudinal analysis of NIST SRE

  2020influential citation
• Deep Neural Networks for Speaker Verfication

  2020cites this paper
• CN-Celeb: multi-genre speaker recognition

  2020influential citation
• Efficient speaker identification using spectral entropy

  2019cites this paper
• An Unsupervised Neural Prediction Framework for Learning Speaker Embeddings Using Recurrent Neural Networks

  2018cites this paper
• Speaker Recognition Using e–Vectors

  2018cites this paper
• Neural Predictive Coding Using Convolutional Neural Networks Toward Unsupervised Learning of Speaker Characteristics

  2018cites this paper
• Advances in Feature Extraction and Modelling for Short Duration Language Identification

  2018cites this paper
• Channel variability synthesis in i-vector speaker recognition

  2017cites this paper
• Nuance - Politecnico di Torino's 2016 NIST Speaker Recognition Evaluation System

  2017cites this paper
• Speaker Identification Using Discriminative Features and Sparse Representation

  2017cites this paper
• Adaptation of PLDA for multi-source text-independent speaker verification

  2017cites this paper
• Speaker Verification Systems Under Various Noise and SNR Conditions

  2017cites this paper
• e-vectors: JFA and i-vectors revisited

  2017cites this paper
• Content-aware local variability vector for speaker verification with short utterance

  2016cites this paper
• Relevance factor of maximum a posteriori adaptation for GMM-NAP-SVM in speaker and language recognition

  2015cites this paper
• Phone-centric local variability vector for text-constrained speaker verification

  2015cites this paper
• I-Vectors for Timbre-Based Music Similarity and Music Artist Classification

  2015cites this paper
• Phoneme dependent inter-session variability reduction for speaker verification

  2015cites this paper
• Speaker Recognition by Machines and Humans: A tutorial review

  2015cites this paper
• Channel adaptation of plda for text-independent speaker verification

  2015influential citation
• A tutorial review ) Speaker Recognition by Machines Humans

  2015cites this paper
• Robust speaker verification using GFCC and joint factor analysis

  2014cites this paper
• Minimum divergence estimation of speaker prior in multi-session PLDA scoring

  2014cites this paper
• JFA-based front ends for speaker recognition

  2014cites this paper
• Joint Factor Analysis for Text-Dependent Speaker Verification

  2014cites this paper
• Automatic recognition of ageing speakers

  2014cites this paper
• Privacy Preserving Speech Processing

  2013cites this paper
• Memory and Computation Trade-Offs for Efficient I-Vector Extraction

  2013cites this paper
• Privacy-preserving speech processing: cryptographic and string-matching frameworks show promise

  2013cites this paper
• Eigenageing compensation for speaker verification

  2013cites this paper
• Multi-System Fusion of Extended Context Prosodic and Cepstral Features for Paralinguistic Speaker Trait Classification

  2012cites this paper
• Approved for public release; distribution unlimited

  2012cites this paper
• Speaker and Language Recognition Techniques

  2012cites this paper
• Memory and computation effective approaches for i - vector extraction

  2012cites this paper
• A joint factor analysis model for handling mismatched recording conditions in forensic automatic speaker recognition

  2012influential citation
• Joint factor analysis for forensic automatic speaker recognition

  2011cites this paper
• Utterance partitioning with acoustic vector resampling for GMM-SVM speaker verification

  2011cites this paper
• Exploitation of phase and vocal excitation modulation features for robust speaker recognition

  2011cites this paper
• Session varaibility compensation in automatic speaker and language recognition

  2011cites this paper
• Subspace Gaussian Mixture Models for Large Vocabulary Speech Recognition

  2010cites this paper
• Local representations and random sampling for speaker verification

  2010cites this paper
• Studies on Model Distance Normalization Approach in Text-independent Speaker Verification

  2009cites this paper
• Underdetermined source separation using speaker subspace models

  2009cites this paper
• Modèles acoustiques à structure temporelle renforcée pour la vérification du locuteur embarquée. (Reinforced temporal structure of acoustic models for speaker recognition)

  2009cites this paper
• Loquendo - Politecnico di Torino's 2008 NIST speaker recognition evaluation system

  2009cites this paper
• Avoiding speaker variability in pronunciation verification of children's disordered speech

  2009cites this paper
• Double Gauss Based Unsupervised Score Normalization in Speaker Verification

  2008cites this paper
• Simplified factor analysis in speaker verification

  2008cites this paper
• Compensation of Nuisance Factors for Speaker and Language Recognition

  2007cites this paper
• Speaker and Session Variability in GMM-Based Speaker Verification

  2007cites this paper
• Joint Factor Analysis Versus Eigenchannels in Speaker Recognition

  2007cites this paper
• Loquendo - Politecnico di torino's 2006 NIST speaker recognition evaluation system

  2007cites this paper
• Channel Factors Compensation in Model and Feature Domain for Speaker Recognition

  2006cites this paper
• Joint Factor Analysis of Speaker and Session Variability: Theory and Algorithms

  2006cites this paper
• Eigenvoice modeling with sparse training data

  2005cites this paper
• Eigen-MLLRs applied to unsupervised speaker enrollment for large vocabulary continuous speech recognition

  2004cites this paper
• Experiments in speaker verification using factor analysis likelihood ratios

  2004cites this paper
• Speaker adaptation using an eigenphone basis

  2004cites this paper
• Disentangling speaker and channel effects in speaker verification

  2004cites this paper