Transforming Binary Uncertainties for Robust Speech Recognition

Recently, several algorithms have been proposed to enhance noisy speech by estimating a binary mask that can be used to select those time-frequency regions of a noisy speech signal that contain more speech energy than noise energy. This binary mask encodes the uncertainty associated with enhanced speech in the linear spectral domain. The use of the cepstral transformation smears the information from the noise dominant time-frequency regions across all the cepstral features. We propose a supervised approach using regression trees to learn the nonlinear transformation of the uncertainty from the linear spectral domain to the cepstral domain. This uncertainty is used by a decoder that exploits the variance associated with the enhanced cepstral features to improve robust speech recognition. Systematic evaluations on a subset of the Aurora4 task using the estimated uncertainty show substantial improvement over the baseline performance across various noise conditions.     

MVA Processing of Speech Features

In this paper, we investigate a technique consisting of mean subtraction, variance normalization and time sequence filtering. Unlike other techniques, it applies auto-regression moving-average (ARMA) filtering directly in the cepstral domain. We call this technique mean subtraction, variance normalization, and ARMA filtering (MVA) post-processing, and speech features with MVA post-processing are called MVA features. Overall, compared to raw features without post-processing, MVA features achieve an error rate reduction of 45% on matched tasks and 65% on mismatched tasks on the Aurora 2.0 noisy speech database, and an average 57% error reduction on the Aurora 3.0 database. These improvements are comparable to the results of much more complicated techniques even though MVA is relatively simple and requires practically no additional computational cost. In this paper, in addition to describing MVA processing, we also present a novel analysis of the distortion of mel-frequency cepstral coefficients and the log energy in the presence of different types of noise. The effectiveness of MVA is extensively investigated with respect to several variations: the configurations used to extract and the type of raw features, the domains where MVA is applied, the filters that are used, the ARMA filter orders, and the causality of the normalization process. Specifically, it is argued and demonstrated that MVA works better when applied to the zeroth-order cepstral coefficient than to log energy, that MVA works better in the cepstral domain, that an ARMA filter is better than either a designed finite impulse response filter or a data-driven filter, and that a five-tap ARMA filter is sufficient to achieve good performance in a variety of settings. We also investigate and evaluate a multi-domain MVA generalization     

Features based on filtering and spectral peaks in autocorrelation domain for robust speech recognition

In this paper, a set of features derived by filtering and spectral peak extraction in autocorrelation domain are proposed. We focus on the effect of the additive noise on speech recognition. Assuming that the channel characteristics and additive noises are stationary, these new features improve the robustness of speech recognition in noisy conditions. In this approach, initially, the autocorrelation sequence of a speech signal frame is computed. Filtering of the autocorrelation of speech signal is carried out in the second step, and then, the short-time power spectrum of speech is obtained from the speech signal through the fast Fourier transform. The power spectrum peaks are then calculated by differentiating the power spectrum with respect to frequency. The magnitudes of these peaks are then projected onto the mel-scale and pass the filter bank. Finally, a set of cepstral coefficients are derived from the outputs of the filter bank. The effectiveness of the new features for speech recognition in noisy conditions will be shown in this paper through a number of speech recognition experiments.A task of multi-speaker isolated-word recognition and another one of multi-speaker continuous speech recognition with various artificially added noises such as factory, babble, car and F16 were used in these experiments. Also, a set of experiments were carried out on Aurora 2 task. Experimental results show significant improvements under noisy conditions in comparison to the results obtained using traditional feature extraction methods. We have also reported the results obtained by applying cepstral mean normalization on the methods to get robust features against both additive noise and channel distortion.     

基于双高斯GMM的特征参数规整及其在语音识别中的应用

对特征参数概率分布的实验分析表明,在有噪声影响的情况下,特征参数通常呈现双峰分布.据此,本文提出了一种新的,基于双高斯的高斯混合模型(Gaussian mixture model,GMM)的特征参数归一化方法,以提高语音识别系统的鲁棒性.该方法采用更为细致的双高斯模型来表达特征参数的累积分布函数(CDF),并依据估计得到的CDF进行参数变换将训练和识别时的特征参数的分布都规整为标准高斯分布,从而提高识别正确率.在Aurora 2和Aurora 3数据库上的实验结果表明,本文提出的方法的性能明显好于传统的倒谱均值规整(Cepstral mean normalization,CMN)和倒谱均值方差规整(Cepstral mean and variance normalization,CMVN)方法,而与非参数化方法-直方图均衡特征规整方法的性能基本相当.     

Non-linear feature extraction for robust speech recognition in stationary and non-stationary noise

An analysis-based non-linear feature extraction approach is proposed, inspired by a model of how speech amplitude spectra are affected by additive noise. Acoustic features are extracted based on the noise-robust parts of speech spectra without losing discriminative information. Two non-linear processing methods, harmonic demodulation and spectral peak-to-valley ratio locking, are designed to minimize mismatch between clean and noisy speech features. A previously studied method, peak isolation [IEEE Transactions on Speech and Audio Processing 5 (1997) 451], is also discussed with this model. These methods do not require noise estimation and are effective in dealing with both stationary and non-stationary noise. In the presence of additive noise, ASR experiments show that using these techniques in the computation of MFCCs improves recognition performance greatly. For the TI46 isolated digits database, the average recognition rate across several SNRs is improved from 60% (using unmodified MFCCs) to 95% (using the proposed techniques) with additive speech-shaped noise. For the Aurora 2 connected digit-string database, the average recognition rate across different noise types, including non-stationary noise background, and SNRs improves from 58% to 80%.     

基于特征参数归一化的鲁棒语音识别方法综述

目前,自动语音识别系统往往会因为环境中复杂因素的影响,造成训练环境和测试环境存在不匹配现象,使得识别系统性能大幅度下降,极大地限制了语音识别技术的应用范围。近年来,很多鲁棒语音识别技术成功地被提出,这些技术的目标都是相同的,主要是提高系统的鲁棒性,进而提高识别率。其中,基于特征的归一化技术简单而有效,常常被作为鲁棒语音识别的首选方法,它主要是通过对特征向量的统计属性、累积密度函数或功率谱的归一化来补偿环境不匹配产生的影响。该文主要对目前主流的归一化方法进行介绍,其中包括倒谱矩归一化方法、直方图均衡化方法以及调频谱归一化方法等。     

语音倒谱特征的研究

语音倒谱特征是语音识别中最常用的特征参数，它表征了人类的听觉特征。该文在研究基于线性预测倒谱和非线性MEL刻度倒谱特征的基础上，研究了LPCC和MFCC参数提取的算法原理及提取算法，提出了一级、二级差分倒谱特征参数的提取算法。识别实验验证了MFCC参数的鲁棒性优于LPCC参数。     

A unified framework of HMM adaptation with joint compensation of additive and convolutive distortions

In this paper, we present our recent development of a model-domain environment robust adaptation algorithm, which demonstrates high performance in the standard Aurora 2 speech recognition task. The algorithm consists of two main steps. First, the noise and channel parameters are estimated using multi-sources of information including a nonlinear environment-distortion model in the cepstral domain, the posterior probabilities of all the Gaussians in speech recognizer, and truncated vector Taylor series (VTS) approximation. Second, the estimated noise and channel parameters are used to adapt the static and dynamic portions (delta and delta–delta) of the HMM means and variances. This two-step algorithm enables joint compensation of both additive and convolutive distortions (JAC). The hallmark of our new approach is the use of a nonlinear, phase-sensitive model of acoustic distortion that captures phase asynchrony between clean speech and the mixing noise.In the experimental evaluation using the standard Aurora 2 task, the proposed Phase-JAC/VTS algorithm achieves 93.32% word accuracy using the clean-trained complex HMM backend as the baseline system for the unsupervised model adaptation. This represents high recognition performance on this task without discriminative training of the HMM system. The experimental results show that the phase term, which was missing in all previous HMM adaptation work, contributes significantly to the achieved high recognition accuracy.     

A Log-Index Weighted Cepstral Distance Measure for Speech Recognition

A log-index weighted cepstral distance measure is proposed and tested in speacker-independent and speaker-dependent isolated word recognition systems using statistic techniques.The weights for the cepstral coefficients of this measure equal the logarithm of the corresponding indices.The experimental results show that this kind of measure works better than any other weighted Euclidean cepstral distance measures on three speech databases.The error rate obtained using this measure is about 1.8 percent for three databases on average,which is a 25% reduction from that obtained using other measures,and a 40% reduction from that obtained using Log Likelihood Ratio(LLR)measure.The experimental results also show that this kind of distance measure woks well in both speaker-dependent and speaker-independent speech recognition systems.     

Speaker recognition under stressed condition

This paper presents the feature analysis and design of compensators for speaker recognition under stressed speech conditions. Any condition that causes a speaker to vary his or her speech production from normal or neutral condition is called stressed speech condition. Stressed speech is induced by emotion, high workload, sleep deprivation, frustration and environmental noise. In stressed condition, the characteristics of speech signal are different from that of normal or neutral condition. Due to changes in speech signal characteristics, performance of the speaker recognition system may degrade under stressed speech conditions. Firstly, six speech features (mel-frequency cepstral coefficients (MFCC), linear prediction (LP) coefficients, linear prediction cepstral coefficients (LPCC), reflection coefficients (RC), arc-sin reflection coefficients (ARC) and log-area ratios (LAR)), which are widely used for speaker recognition, are analyzed for evaluation of their characteristics under stressed condition. Secondly, Vector Quantization (VQ) classifier and Gaussian Mixture Model (GMM) are used to evaluate speaker recognition results with different speech features. This analysis help select the best feature set for speaker recognition under stressed condition. Finally, four VQ based novel compensation techniques are proposed and evaluated for improvement of speaker recognition under stressed condition. The compensation techniques are speaker and stressed information based compensation (SSIC), compensation by removal of stressed vectors (CRSV), cepstral mean normalization (CMN) and combination of MFCC and sinusoidal amplitude (CMSA) features. Speech data from SUSAS database corresponding to four different stressed conditions, Angry, Lombard, Question and Neutral, are used for analysis of speaker recognition under stressed condition.     

Preliminary experiments in speaker verification using time-dependent largest Lyapunov exponents

The characterization of a speech signal using non-linear dynamical features has been the focus of intense research lately. In this work, the results obtained with time-dependent largest Lyapunov exponents (TDLEs) in a text-dependent speaker verification task are reported. The baseline system used Gaussian mixture models (GMMs), obtained from the adaptation of a universal background model (UBM), for the speaker voice models. Sixteen cepstral and 16 delta cepstral features were used in the experiments, and it is shown how the addition of TDLEs can improve the system’s accuracy. Cepstral mean subtraction was applied to all features in the tests for channel equalization, and silence frames were discarded. The corpus used, obtained from a subset of the Center for Spoken Language Understanding (CSLU) Speaker Recognition corpus, consisted of telephone speech from 91 different speakers.     

Environmental robust speech and speaker recognition through multi-channel histogram equalization

Feature statistics normalization in the cepstral domain is one of the most performing approaches for robust automaticspeech and speaker recognition in noisy acoustic scenarios: feature coefficients are normalized by using suitable linear or nonlinear transformations in order to match the noisy speech statistics to the clean speech one. Histogram equalization (HEQ) belongs to such a category of algorithms and has proved to be effective on purpose and therefore taken here as reference.In this paper the presence of multi-channel acoustic channels is used to enhance the statistics modeling capabilities of the HEQ algorithm, by exploiting the availability of multiple noisy speech occurrences, with the aim of maximizing the effectiveness of the cepstra normalization process. Computer simulations based on the Aurora 2 database in speech and speaker recognition scenarios have shown that a significant recognition improvement with respect to the single-channel counterpart and other multi-channel techniques can be achieved confirming the effectiveness of the idea. The proposed algorithmic configuration has also been combined with the kernel estimation technique in order to further improve the speech recognition performances.     

Pseudo pitch synchronous analysis of speech with applications to speaker recognition

The fine spectral structure related to pitch information is conveyed in Mel cepstral features, with variations in pitch causing variations in the features. For speaker recognition systems, this phenomenon, known as "pitch mismatch" between training and testing, can increase error rates. Likewise, pitch-related variability may potentially increase error rates in speech recognition systems for languages such as English in which pitch does not carry phonetic information. In addition, for both speech recognition and speaker recognition systems, the parsing of the raw speech signal into frames is traditionally performed using a constant frame size and a constant frame offset, without aligning the frames to the natural pitch cycles. As a result the power spectral estimation that is done as part of the Mel cepstral computation may include artifacts. Pitch synchronous methods have addressed this problem in the past, at the expense of adding some complexity by using a variable frame size and/or offset. This paper introduces Pseudo Pitch Synchronous (PPS) signal processing procedures that attempt to align each individual frame to its natural cycle and avoid truncation of pitch cycles while still using constant frame size and frame offset, in an effort to address the above problems. Text independent speaker recognition experiments performed on NIST speaker recognition tasks demonstrate a performance improvement when the scores produced by systems using PPS are fused with traditional speaker recognition scores. In addition, a better distribution of errors across trials may be obtained for similar error rates, and some insight regarding of role of the fundamental frequency in speaker recognition is revealed. Speech recognition experiments run on the Aurora-2 noisy digits task also show improved robustness and better accuracy for extremely low signal-to-noise ratio (SNR) data.     

Robust Feature Extraction for Continuous Speech Recognition Using the MVDR Spectrum Estimation Method

This paper describes a robust feature extraction technique for continuous speech recognition. Central to the technique is the minimum variance distortionless response (MVDR) method of spectrum estimation. We consider incorporating perceptual information in two ways: 1) after the MVDR power spectrum is computed and 2) directly during the MVDR spectrum estimation. We show that incorporating perceptual information directly into the spectrum estimation improves both robustness and computational efficiency significantly. We analyze the class separability and speaker variability properties of the features using a Fisher linear discriminant measure and show that these features provide better class separability and better suppression of speaker-dependent information than the widely used mel frequency cepstral coefficient (MFCC) features. We evaluate the technique on four different tasks: an in-car speech recognition task, the Aurora-2 matched task, the Wall Street Journal (WSJ) task, and the Switchboard task. The new feature extraction technique gives lower word-error-rates than the MFCC and perceptual linear prediction (PLP) feature extraction techniques in most cases. Statistical significance tests reveal that the improvement is most significant in high noise conditions. The technique thus provides improved robustness to noise without sacrificing performance in clean conditions     

Local spectral variability features for speaker verification

Speaker verification techniques neglect the short-time variation in the feature space even though it contains speaker related attributes. We propose a simple method to capture and characterize this spectral variation through the eigenstructure of the sample covariance matrix. This covariance is computed using sliding window over spectral features. The newly formulated feature vectors representing local spectral variations are used with classical and state-of-the-art speaker recognition systems. Results on multiple speaker recognition evaluation corpora reveal that eigenvectors weighted with their normalized singular values are useful in representing local covariance information. We have also shown that local variability features can be extracted using mel frequency cepstral coefficients (MFCCs) as well as using three recently developed features: frequency domain linear prediction (FDLP), mean Hilbert envelope coefficients (MHECs) and power-normalized cepstral coefficients (PNCCs). Since information conveyed in the proposed feature is complementary to the standard short-term features, we apply different fusion techniques. We observe considerable relative improvements in speaker verification accuracy in combined mode on text-independent (NIST SRE) and text-dependent (RSR2015) speech corpora. We have obtained up to 12.28% relative improvement in speaker recognition accuracy on text-independent corpora. Conversely in experiments on text-dependent corpora, we have achieved up to 40% relative reduction in EER. To sum up, combining local covariance information with the traditional cepstral features holds promise as an additional speaker cue in both text-independent and text-dependent recognition.     

Robust feature extraction based on an asymmetric level-dependent auditory filterbank and a subband spectrum enhancement technique

In this paper we introduce a robust feature extractor, dubbed as robust compressive gammachirp filterbank cepstral coefficients (RCGCC), based on an asymmetric and level-dependent compressive gammachirp filterbank and a sigmoid shape weighting rule for the enhancement of speech spectra in the auditory domain. The goal of this work is to improve the robustness of speech recognition systems in additive noise and real-time reverberant environments. As a post processing scheme we employ a short-time feature normalization technique called short-time cepstral mean and scale normalization (STCMSN), which, by adjusting the scale and mean of cepstral features, reduces the difference of cepstra between the training and test environments. For performance evaluation, in the context of speech recognition, of the proposed feature extractor we use the standard noisy AURORA-2 connected digit corpus, the meeting recorder digits (MRDs) subset of the AURORA-5 corpus, and the AURORA-4 LVCSR corpus, which represent additive noise, reverberant acoustic conditions and additive noise as well as different microphone channel conditions, respectively. The ETSI advanced front-end (ETSI-AFE), the recently proposed power normalized cepstral coefficients (PNCC), conventional MFCC and PLP features are used for comparison purposes. Experimental speech recognition results demonstrate that the proposed method is robust against both additive and reverberant environments. The proposed method provides comparable results to that of the ETSI-AFE and PNCC on the AURORA-2 as well as AURORA-4 corpora and provides considerable improvements with respect to the other feature extractors on the AURORA-5 corpus.     

A noise-robust front-end for distributed speech recognition in mobile communications

This paper investigates a new front-end processing that aims at improving the performance of speech recognition in noisy mobile environments. This approach combines features based on conventional Mel-cepstral Coefficients (MFCCs), Line Spectral Frequencies (LSFs) and formant-like (FL) features to constitute robust multivariate feature vectors. The resulting front-end constitutes an alternative to the DSR-XAFE (XAFE: eXtended Audio Front-End) available in GSM mobile communications. Our results showed that for highly noisy speech, using the paradigm that combines these spectral cues leads to a significant improvement in recognition accuracy on the Aurora 2 task.     

ELM speaker identification for limited dataset using multitaper based MFCC and PNCC features with fusion score

In current scenario, speaker recognition under noisy condition is the major challenging task in the area of speech processing. Due to noise environment there is a significant degradation in the system performance. The major aim of the proposed work is to identify the speaker’s under clean and noise background using limited dataset. In this paper, we proposed a multitaper based Mel frequency cepstral coefficients (MFCC) and power normalization cepstral coefficients (PNCC) techniques with fusion strategies. Here, we used MFCC and PNCC techniques with different multitapers to extract the desired features from the obtained speech samples. Then, cepstral mean and variance normalization (CMVN) and Feature warping (FW) are the two techniques applied to normalize the obtained features from both the techniques. Furthermore, as a system model low dimension i-vector model is used and also different fusion score strategies like mean, maximum, weighted sum, cumulative and concatenated fusion techniques are utilized. Finally extreme learning machine (ELM) is used for classification in order to increase the system identification accuracy (SIA) intern which is having a single layer feedforward neural network with less complexity and time consuming compared to other neural networks. TIMIT and SITW 2016 are the two different databases are used to evaluate the proposed system under limited data of these databases. Both clean and noisy backgrounds conditions are used to check the SIA.

     

Robust speech recognition using spatial-temporal feature distribution characteristics

Histogram equalization (HEQ) is one of the most efficient and effective techniques that have been used to reduce the mismatch between training and test acoustic conditions. However, most of the current HEQ methods are merely performed in a dimension-wise manner and without allowing for the contextual relationships between consecutive speech frames. In this paper, we present several novel HEQ approaches that exploit spatial-temporal feature distribution characteristics for speech feature normalization. The automatic speech recognition (ASR) experiments were carried out on the Aurora-2 standard noise-robust ASR task. The performance of the presented approaches was thoroughly tested and verified by comparisons with the other popular HEQ methods. The experimental results show that for clean-condition training, our approaches yield a significant word error rate reduction over the baseline system, and also give competitive performance relative to the other HEQ methods compared in this paper.