Efficient Cepstral Normalization for Robust Speech Recognition

In this paper we describe and compare the performance of a series of cepstrum-based procedures that enable the CMU SPHINX-II speech recognition system to maintain a high level of recognition accuracy over a wide variety of acoustical environments. We describe the MFCDCN algorithm, an environment-independent extension of the efficient SDCN and FCDCN algorithms developed previously. We compare the performance of these algorithms with the very simple RASTA and cepstral mean normalization procedures, describing the performance of these algorithms in the context of the 1992 DARPA CSR evaluation using secondary microphones, and in the DARPA stress-test evaluation.

• Publication year

  1993

• Venue

  Human Language Technology - The Baltic Perspectiv

• Publication date

  1993-03-21

• Fields of study

  Computer Science

• Identifiers
  DOI 10.3115/1075671.1075688
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Speech recognition in adverse environments

  1994cited by this paper
• Comparative Experiments on Large Vocabulary Speech Recognition

  1993cited by this paper
• An Overview of the SPHINX-II Speech Recognition System

  1993influential reference
• RASTA-PLP speech analysis technique

  1992cited by this paper
• Reduced Channel Dependence for Speech Recognition

  1992cited by this paper
• Efficient joint compensation of speech for the effects of additive noise and linear filtering

  1992cited by this paper
• Speech recognition in adverse environments

  1991cited by this paper
• Robust speech recognition by normalization of the acoustic space

  1991influential reference
• Acoustical and environmental robustness in automatic speech recognition

  1991influential reference
• Estimation using log-spectral-distance criterion for noise-robust speech recognition

  1990cited by this paper
• Environmental robustness in automatic speech recognition

  1990influential reference
• Comparison of Parametric Representations for Monosyllabic Word Recognition in Continuously Spoken Se

  1980cited by this paper
• Suppression of acoustic noise in speech using spectral subtraction

  1979cited by this paper
• Blind deconvolution through digital signal processing

  1975cited by this paper

• Exploring Dynamic Parameters for Vietnamese Gender-Independent ASR

  2025cites this paper
• A review of social background profiling of speakers from speech accents

  2024cites this paper
• Cardiac anomaly detection considering an additive noise and convolutional distortion model of heart sound recordings

  2022cites this paper
• Speaker Adaptation on Articulation and Acoustics for Articulation-to-Speech Synthesis

  2022cites this paper
• Heart Sound Classification Considering Additive Noise and Convolutional Distortion

  2021cites this paper
• Recognition-Synthesis Based Non-Parallel Voice Conversion with Adversarial Learning

  2020cites this paper
• On Front-end Gain Invariant Modeling for Wake Word Spotting

  2020cites this paper
• Deep Scattering Power Spectrum Features for Robust Speech Recognition

  2020cites this paper
• Adaptive Speaker Normalization for CTC-Based Speech Recognition

  2020cites this paper
• Frustratingly Easy Noise-aware Training of Acoustic Models

  2020influential citation
• The preliminary study of robust speech feature extraction based on maximizing the accuracy of states in deep acoustic models

  2020cites this paper
• ELM speaker identification for limited dataset using multitaper based MFCC and PNCC features with fusion score

  2020cites this paper
• Kaleidoscope: An Efficient, Learnable Representation For All Structured Linear Maps

  2020cites this paper
• Title Comparative evaluation of modulation-transfer-function-based blind restoration of sub-band power envelopes of speech as a front-end processor for automatic speech recognition systems

  2019influential citation
• Speech recognition based strategies for on-line Computer Assisted Language Learning (CALL) systems in Basque

  2019cites this paper
• The JHU ASR System for VOiCES from a Distance Challenge 2019

  2019cites this paper
• Automatic Speech Recognition System Based on Hybrid Feature Extraction Techniques Using TEO-PWP for in Real Noisy Environment

  2019cites this paper
• Hizketa-ezagutzan oinarritutako estrategiak, euskarazko online OBHI (Ordenagailu Bidezko Hizkuntza Ikaskuntza) sistemetarako

  2019cites this paper
• Research of Window Function Influence on the Result of Arabic Speech Automatic Recognition

  2019cites this paper
• End-to-end Anchored Speech Recognition

  2019cites this paper
• Benchmarking Neural Network Robustness to Common Corruptions and Perturbations

  2019cites this paper
• Transfer Linear Subspace Learning for Cross-Corpus Speech Emotion Recognition

  2019cites this paper
• Benchmarking Neural Network Robustness to Common Corruptions and Surface Variations

  2018cites this paper
• A comparative study of deep neural network based Punjabi-ASR system

  2018influential citation
• Chapter 4 Deep Neural Networks

  2018cites this paper
• An on-line VAD based on Multi-Normalisation Scoring (MNS) of observation likelihoods

  2018cites this paper
• Algorithm to demodulate an electromyogram signal modulated by essential tremor

  2017cites this paper
• Multi-modal and deep learning for robust speech recognition

  2017cites this paper
• Spatial Cepstrum as a Spatial Feature Using a Distributed Microphone Array for Acoustic Scene Analysis

  2017influential citation
• Acoustic scene classification using asynchronous multichannel observations with different lengths

  2017cites this paper
• Filter Bank Energy Based Malayalam Speech Segmentation and Recognition

  2017cites this paper
• Large vocabulary continuous speech recognition for the transcription of Catalan broadcast news and conversations : towards analysis and modelling of acoustic reduction in spontaneous speech

  2017cites this paper
• The Use of Locally Normalized Cepstral Coefficients (LNCC) to Improve Speaker Recognition Accuracy in Highly Reverberant Rooms

  2016cites this paper
• Optimizing feature extraction techniques constituting phone based modelling on connected words for Punjabi automatic speech recognition

  2016cites this paper
• Power-Normalized Cepstral Coefficients (PNCC) for Punjabi Automatic Speech Recognition using phone based modelling in HTK

  2016cites this paper
• A Feature Normalisation Technique for PLLR Based Language Identification Systems

  2016cites this paper
• Recapitulation on Audio-Visual features for Speech Recognition

  2016cites this paper
• Anchored Speech Detection

  2016cites this paper
• An analysis of two common reference points for EEGS

  2016cites this paper
• РАЗРАБОТКА СИСТЕМЫ АВТОМАТИЧЕСКОГО РАСПОЗНАВАНИЯ РЕЧИДЛЯ ЕГИПЕТСКОГО ДИАЛЕКТА АРАБСКОГО ЯЗЫКАВ ТЕЛЕФОННОМ КАНАЛЕ

  2016cites this paper
• DEVELOPMENT OF AUTOMATED SPEECH RECOGNITION SYSTEM FOR EGYPTIAN ARABIC PHONE CONVERSATIONS

  2016cites this paper
• Design and Evaluation of Statistical Parametric Techniques in Expressive Text-To-Speech: Emotion and Speaking Styles Transplantation

  2016cites this paper
• Taylor series expansion of psychoacoustic corruption function for noise robust speech recognition

  2016cites this paper
• Hidden Markov Models and the Variants

  2015cites this paper
• Noise Reduction for ILSE Interviews

  2015cites this paper
• Speech selection and environmental adaptation for asynchronous speech recognition

  2015cites this paper
• Noise Reduction for ILSE Interviews Asunción Pastrana

  2015cites this paper
• A perceptually-motivated low-complexity instantaneous linear channel normalization technique applied to speaker verification

  2015cites this paper
• Deep Neural Networks

  2015cites this paper
• Spatial-feature-based acoustic scene analysis using distributed microphone array

  2015cites this paper
• Combination of bottleneck feature extraction and dereverberation for distant-talking speech recognition

  2015cites this paper
• Frame-level features conveying phonetic information for language and speaker recognition

  2015cites this paper
• Stereo-based histogram equalization for robust speech recognition

  2015influential citation
• Phase Autocorrelation Bark Wavelet Transform (PACWT) Features for Robust Speech Recognition

  2015cites this paper
• Perceptual compensation for reverberation in human listeners and machines

  2015cites this paper
• Deep neural network-based bottleneck feature and denoising autoencoder-based dereverberation for distant-talking speaker identification

  2015cites this paper
• Noise Robust Speech Parameterization using Relative Spectra and Auditory Filterbank

  2015cites this paper
• Environment-dependent denoising autoencoder for distant-talking speech recognition

  2015cites this paper
• Features and Model Adaptation Techniques for Robust Speech Recognition: A Review

  2015cites this paper
• CoHear: Co-location Detection of Mobile Devices using Ambient Sound

  2015cites this paper
• Regularized minimum variance distortionless response-based cepstral features for robust continuous speech recognition

  2015cites this paper
• Robust feature extraction based on an asymmetric level-dependent auditory filterbank and a subband spectrum enhancement technique

  2014cites this paper
• Chapter 3: Hidden Markov Models and the Variants

  2014cites this paper
• Acoustic Model Structuring for Improving Automatic Speech Recognition Performance. (Structuration du modèle acoustique pour améliorer les performances de la reconnaissance automatique de la parole)

  2014cites this paper
• Distant-talking speaker identification by generalized spectral subtraction-based dereverberation and its efficient computation

  2014influential citation
• Single-channel Dereverberation for Distant-Talking Speech Recognition by Combining Denoising Autoencoder and Temporal Structure Normalization

  2014cites this paper
• Regularized Estimation of Gaussian Mixture Models for SVM Based Speaker Recognition

  2014cites this paper
• Development of a filtering algorithm to demodulate electromyogram signal of essential tremor patients

  2014cites this paper
• Mean normalization of power function based cepstral coefficients for robust speech recognition in noisy environment

  2014cites this paper
• Feature-based noise robust speech recognition on an Indonesian language automatic speech recognition system

  2014cites this paper
• Denoising autoencoder and environment adaptation for distant-talking speech recognition with asynchronous speech recording

  2014cites this paper
• Robust feature extractors for continuous speech recognition

  2014cites this paper
• Dynamic automatic noisy speech recognition system: (DANSR) (Dynamische automatische verrauschte Spracherkennung)

  2014cites this paper
• Robustness of Speech Recognition System of Isolated Speech in Macedonian

  2014cites this paper
• An approach to vowel recognition using 2DDWT based visual information of the lip region

  2014cites this paper
• Optimizing acoustic features for source cell-phone recognition using speech signals

  2013cites this paper
• Connected-digits recognition for an under-resourced language using Hidden Markov Models

  2013influential citation
• Aging speech recognition with speaker adaptation techniques: Study on medium vocabulary continuous Bengali speech

  2013cites this paper
• Implementation of the Intelligent Voice System for Kazakh

  2013cites this paper
• Security Issues In Speech Watermarking for Information Transmission

  2013cites this paper
• Towards Building an Intelligent Voice System for Kazakh: Acoustic Database and System Design

  2013cites this paper
• The Reduction of Speech Characteristic Vector Using PSO Algorithm and the Evaluation of the Effectiveness of Different Speech Characteristic in Recognition of Persian Language State

  2012cites this paper
• Analysis of EMG signals of patients with essential tremor focusing on the change of tremor frequency

  2012cites this paper
• 0 Multidimensional Features Extraction Methods in Frequency Domain

  2012cites this paper
• Automatic Language Identification in speech streams

  2012cites this paper
• Multi-level acoustic modeling for automatic speech recognition

  2012cites this paper
• Robust Feature Extraction for Speech Recognition by Enhancing Auditory Spectrum

  2012cites this paper
• Técnicas de reconocimiento robusto de la voz basadas en el pitch

  2012cites this paper
• Tremor frequency based filter to extract voluntary movement of patients with essential tremor

  2012cites this paper
• A Double-Ended Interference Assessing System for Short Wave Transmission

  2012influential citation
• On the use of phone log-likelihood ratios as features in spoken language recognition

  2012cites this paper
• A Double-Ended Interference Assessing System for Short Wave Transmission

  2012cites this paper
• Effects of the Phonological Contents and Transmission Channels on Forensic Speaker Recognition

  2012cites this paper
• Speaker diarization and tracking in multiple-sensor environments

  2012cites this paper
• Multidimensional Features Extraction Methods in Frequency Domain

  2012cites this paper
• Distant-Talking Speech Recognition Based on Spectral Subtraction by Multi-Channel LMS Algorithm

  2011cites this paper
• An Overview of Speaker Identification: Accuracy and Robustness Issues

  2011cites this paper
• Signal and feature domain enhancement approaches for robust speech recognition

  2011cites this paper
• Development of robotic upper limb orthosis with tremor suppressiblity and elbow joint movability

  2011cites this paper
• Sub-band temporal modulation envelopes and their normalization for automatic speech recognition in reverberant environments

  2011cites this paper