Using geometric spectral subtraction approach for feature extraction for DSR front-end Arabic system

Noise robustness and Arabic language are still considered as the main challenges for speech recognition over mobile environments. This paper contributed to these trends by proposing a new robust Distributed Speech Recognition (DSR) system for Arabic language. A speech enhancement algorithm was applied to the noisy speech as a robust front-end pre-processing stage to improve the recognition performance. While an isolated Arabic word engine was designed, and developed using HMM Model to perform the recognition process at the back-end. To test the engine, several conditions including clean, noisy and enhanced noisy speech were investigated together with speaker dependent and speaker independent tasks. With the experiments carried out on noisy database, multi-condition training outperforms the clean training mode in all noise types in terms of recognition rate. The results also indicate that using the enhancement method increases the DSR accuracy of our system under severe noisy conditions especially at low SNR down to 10 dB.     

Voice activity detection algorithm using nonlinear spectral weights,hangover and hangbefore criteria

This paper introduces a nonlinear function into the frequency spectrum that improves the detection of vowels, diphthongs, and semivowels within the speech signal. The lower efficiency of consonant detection was solved by implementing the hangover and hangbefore criteria. This paper presents a procedure for faster definition of those optimal constants used by hangover and hangbefore criteria. A nonlinearly changed frequency spectrum is used in the proposed GMM (Gaussian Mixture Model) based VAD (Voice Activity Detection) algorithm. Comparative tests between the proposed VAD algorithm and seven other VAD algorithms were made on the Aurora 2 database. The experiments were based on frame error detection and on speech recognition performance for two types of acoustic training modes (multi-condition and clean only). The lowest average percentage of frame errors was obtained by the proposed VAD algorithm, which also achieved positive improvement in the speech recognition performance for both types of acoustic training modes.     

Feature enhancement by deep LSTM networks for ASR in reverberant multisource environments

This article investigates speech feature enhancement based on deep bidirectional recurrent neural networks. The Long Short-Term Memory (LSTM) architecture is used to exploit a self-learnt amount of temporal context in learning the correspondences of noisy and reverberant with undistorted speech features. The resulting networks are applied to feature enhancement in the context of the 2013 2nd Computational Hearing in Multisource Environments (CHiME) Challenge track 2 task, which consists of the Wall Street Journal (WSJ-0) corpus distorted by highly non-stationary, convolutive noise. In extensive test runs, different feature front-ends, network training targets, and network topologies are evaluated in terms of frame-wise regression error and speech recognition performance. Furthermore, we consider gradually refined speech recognition back-ends from baseline ‘out-of-the-box’ clean models to discriminatively trained multi-condition models adapted to the enhanced features. In the result, deep bidirectional LSTM networks processing log Mel filterbank outputs deliver best results with clean models, reaching down to 42% word error rate (WER) at signal-to-noise ratios ranging from −6 to 9 dB (multi-condition CHiME Challenge baseline: 55% WER). Discriminative training of the back-end using LSTM enhanced features is shown to further decrease WER to 22%. To our knowledge, this is the best result reported for the 2nd CHiME Challenge WSJ-0 task yet.     

Recognition of Arabic (Indian) bank check digits using log-gabor filters

In this paper we present a technique for the automatic recognition of Arabic (Indian) bank check digits based on features extracted by using the Log Gabor filters. The digits are classified by using the K-Nearest Neighbor (K-NN), Hidden Markov Models (HMM) and Support Vector Machines (SVM) classifiers. An extensive experimentation is conducted on the CENPARMI data, a database consisting of 7390 samples of Arabic (Indian) digits for training and 3035 samples for testing extracted from real bank checks. The data is normalized to a height of 64 pixels, maintaining the aspect ratio. Log Gabor filters with several scales and orientations are used. In addition, the filtered images are segmented into different region sizes for feature extraction. Recognition rates of 98.95%, 98.75%, 98.62%, 97.21% and 94.43% are achieved with SVM, 1-NN, 3-NN, HMM and NM classifiers, respectively. These results significantly outperform published work using the same database. The misclassified digits are evaluated subjectively and results indicate that human subjects misclassified 1/3 of these digits. The experimental results, including the subjective evaluation of misclassified digits, indicate the effectiveness of the selected Log Gabor filters parameters, the implemented image segmentation technique, and extracted features for practical recognition of Arabic (Indian) digits.     

Farsi/Arabic handwritten digit recognition based on ensemble of SVD classifiers and reliable multi-phase PSO combination rule

The problem of handwritten digit recognition has long been an open problem in the field of pattern classification and of great importance in industry. The heart of the problem lies within the ability to design an efficient algorithm that can recognize digits written and submitted by users via a tablet, scanner, and other digital devices. From an engineering point of view, it is desirable to achieve a good performance within limited resources. To this end, we have developed a new approach for handwritten digit recognition that uses a small number of patterns for training phase. To improve the overall performance achieved in classification task, the literature suggests combining the decision of multiple classifiers rather than using the output of the best classifier in the ensemble; so, in this new approach, an ensemble of classifiers is used for the recognition of handwritten digit. The classifiers used in proposed system are based on singular value decomposition (SVD) algorithm. The experimental results and the literature show that the SVD algorithm is suitable for solving sparse matrices such as handwritten digit. The decisions obtained by SVD classifiers are combined by a novel proposed combination rule which we named reliable multi-phase particle swarm optimization. We call the method “Reliable” because we have introduced a novel reliability parameter which is applied to tackle the problem of PSO being trapped in local minima. In comparison with previous methods, one of the significant advantages of the proposed method is that it is not sensitive to the size of training set. Unlike other methods, the proposed method uses just 15 % of the dataset as a training set, while other methods usually use (60–75) % of the whole dataset as the training set. To evaluate the proposed method, we tested our algorithm on Farsi/Arabic handwritten digit dataset. What makes the recognition of the handwritten Farsi/Arabic digits more challenging is that some of the digits can be legally written in different shapes. Therefore, 6000 hard samples (600 samples per class) are chosen by K-nearest neighbor algorithm from the HODA dataset which is a standard Farsi/Arabic digit dataset. Experimental results have shown that the proposed method is fast, accurate, and robust against the local minima of PSO. Finally, the proposed method is compared with state of the art methods and some ensemble classifier based on MLP, RBF, and ANFIS with various combination rules.     

Hybrid continuous speech recognition systems by HMM,MLP and SVM: a comparative study

This paper presents a new hybrid method for continuous Arabic speech recognition based on triphones modelling. To do this, we apply Support Vectors Machine (SVM) as an estimator of posterior probabilities within the Hidden Markov Models (HMM) standards. In this work, we describe a new approach of categorising Arabic vowels to long and short vowels to be applied on the labeling phase of speech signals. Using this new labeling method, we deduce that SVM/HMM hybrid model is more efficient then HMMs standards and the hybrid system Multi-Layer Perceptron (MLP) with HMM. The obtained results for the Arabic speech recognition system based on triphones are 64.68 % with HMMs, 72.39 % with MLP/HMM and 74.01 % for SVM/HMM hybrid model. The WER obtained for the recognition of continuous speech by the three systems proves the performance of SVM/HMM by obtaining the lowest average for 4 tested speakers 11.42 %.     

Comparing ANN to HMM in implementing limited Arabic vocabulary ASR systems

In this paper we investigated Artificial Neural Networks (ANN) based Automatic Speech Recognition (ASR) by using limited Arabic vocabulary corpora. These limited Arabic vocabulary subsets are digits and vowels carried by specific carrier words. In addition to this, Hidden Markov Model (HMM) based ASR systems are designed and compared to two ANN based systems, namely Multilayer Perceptron (MLP) and recurrent architectures, by using the same corpora. All systems are isolated word speech recognizers. The ANN based recognition system achieved 99.5% correct digit recognition. On the other hand, the HMM based recognition system achieved 98.1% correct digit recognition. With vowels carrier words, the MLP and recurrent ANN based recognition systems achieved 92.13% and 98.06, respectively, correct vowel recognition; but the HMM based recognition system achieved 91.6% correct vowel recognition.     

基于发音特征的声效相关鲁棒语音识别算法

针对声效(VE)相关的语音识别鲁棒性问题,提出了基于多模型框架的语音识别算法.首先,分析了不同声效模式下语音信号的声学特性以及声效变化对语音识别精度的影响;然后,提出了基于高斯混合模型(GMM)的声效模式检测方法;最后,根据声效检测的结果,训练专门的声学模型用于耳语音识别,而将发音特征与传统的谱特征一起用于其余4种声效模式的语音识别.基于孤立词识别的实验结果显示,采用所提方法后语音识别准确率有了明显的提高:与基线系统相比,所提方法5种声效的平均字错误率降低了26.69%;与声学模型混合语料训练方法相比,平均字错误率降低了14.51%;与最大似然线性回归(MLLR)自适应方法相比,平均字错误率降低了15.30%.实验结果表明:与传统谱特征相比发音特征对于声效变化更具鲁棒性,而多模型框架是解决声效相关的语音识别鲁棒性问题的有效方法.     

Stochastic feature compensation methods for speaker verification in noisy environments

This paper explores the significance of stereo-based stochastic feature compensation (SFC) methods for robust speaker verification (SV) in mismatched training and test environments. Gaussian Mixture Model (GMM)-based SFC methods developed in past has been solely restricted for speech recognition tasks. Application of these algorithms in a SV framework for background noise compensation is proposed in this paper. A priori knowledge about the test environment and availability of stereo training data is assumed. During the training phase, Mel frequency cepstral coefficient (MFCC) features extracted from a speaker's noisy and clean speech utterance (stereo data) are used to build front end GMMs. During the evaluation phase, noisy test utterances are transformed on the basis of a minimum mean squared error (MMSE) or maximum likelihood (MLE) estimate, using the target speaker GMMs. Experiments conducted on the NIST-2003-SRE database with clean speech utterances artificially degraded with different types of additive noises reveal that the proposed SV systems strictly outperform baseline SV systems in mismatched conditions across all noisy background environments.     

基于深度神经网络的语音识别系统研究

语音识别是人机交互模式识别领域的一个重要课题,构建了一种基于深度神经网络的语音识别系统,使用了抗噪对比散度法和抗噪最小平方误差法对模型进行无监督训练;使用了均值归一化进行模型优化, 提高了网络对训练集的拟合度,并且降低了语音识别的错误率;使用多状态激活函数进行了模型优化,这不仅使得不带噪测试和带噪声测试的语音识别错误率进一步下降,并能在一定程度上减轻过拟合现象;并通过奇异值分解和重构的方法对模型进行了降维。实验结果表明,此系统可以在不影响语音识别错误率的基础上极大地降低系统的复杂性。     

A Study of Variable-Parameter Gaussian Mixture Hidden Markov Modeling for Noisy Speech Recognition

To improve recognition performance in noisy environments, multicondition training is usually applied in which speech signals corrupted by a variety of noise are used in acoustic model training. Published hidden Markov modeling of speech uses multiple Gaussian distributions to cover the spread of the speech distribution caused by noise, which distracts the modeling of speech event itself and possibly sacrifices the performance on clean speech. In this paper, we propose a novel approach which extends the conventional Gaussian mixture hidden Markov model (GMHMM) by modeling state emission parameters (mean and variance) as a polynomial function of a continuous environment-dependent variable. At the recognition time, a set of HMMs specific to the given value of the environment variable is instantiated and used for recognition. The maximum-likelihood (ML) estimation of the polynomial functions of the proposed variable-parameter GMHMM is given within the expectation-maximization (EM) framework. Experiments on the Aurora 2 database show significant improvements of the variable-parameter Gaussian mixture HMMs compared to the conventional GMHMMs     

Some performance benchmarks for isolated work speech recognition systems

The performance of isolated word speech recognition system has steadily improved over time as we learn more about how to represent the significant events in speech, and how to capture these events via appropriate analysis procedures and training algorithms. In particular, algorithms based on both template matching (via dynamic time warping (DTW) procedures) and hidden Markov models (HMMs) have been developed which yield high accuracy on several standard vocabularies, including the 10 digits (zero to nine) and the set of 26 letters of the English alphabet (A-Z). Results are given showing currently attainable performance of a laboratory system for both template-based (DTW) and HMM-based recognizers, operating in both speaker trained and speaker independent modes, on the digits and the alphabet vocabularies using telephone recordings. We show that the average error rates of these systems, on standard vocabularies, are significantly lower than those reported several years back on the exact same databases, thereby reflecting the progress which has been made in all aspects of the speech recognition process.     

基于数据模拟和HMM自适应的电话语音识别研究

该文研究了基于数据模拟方法和HMM(隐马尔科夫模型)自适应的电话信道条件下语音识别问题。模拟数据模仿了纯净语音在不同电话信道条件下的语音行为。各基线系统的HMM模型分别由纯净语音和模拟语音训练而成。语音识别实验评估了各基线系统HMM模型在采用MLLR算法(最大似然线性回归)做无监督式自适应前后的识别性能。实验证明,由纯净语音转换生成的模拟语音有效地减小了训练语音和测试语音声学性质的不匹配,很大程度上提高了电话语音识别率。基线模型的自适应结果显示模拟数据的自适应性能比纯净语音自适应的性能最大提高达到9.8%,表明了电话语音识别性能的进一步改善和系统稳健性的提高。     

Phonetically rich and balanced text and speech corpora for Arabic language

This paper describes the preparation, recording, analyzing, and evaluation of a new speech corpus for Modern Standard Arabic (MSA). The speech corpus contains a total of 415 sentences recorded by 40 (20 male and 20 female) Arabic native speakers from 11 different Arab countries representing three major regions (Levant, Gulf, and Africa). Three hundred and sixty seven sentences are considered as phonetically rich and balanced, which are used for training Arabic Automatic Speech Recognition (ASR) systems. The rich characteristic is in the sense that it must contain all phonemes of Arabic language, whereas the balanced characteristic is in the sense that it must preserve the phonetic distribution of Arabic language. The remaining 48 sentences are created for testing purposes, which are mostly foreign to the training sentences and there are hardly any similarities in words. In order to evaluate the speech corpus, Arabic ASR systems were developed using the Carnegie Mellon University (CMU) Sphinx 3 tools at both training and testing/decoding levels. The speech engine uses 3-emitting state Hidden Markov Models (HMM) for tri-phone based acoustic models. Based on experimental analysis of about 8?h of training speech data, the acoustic model is best using continuous observation’s probability model of 16 Gaussian mixture distributions and the state distributions were tied to 500 senones. The language model contains uni-grams, bi-grams, and tri-grams. For same speakers with different sentences, Arabic ASR systems obtained average Word Error Rate (WER) of 9.70%. For different speakers with same sentences, Arabic ASR systems obtained average WER of 4.58%, whereas for different speakers with different sentences, Arabic ASR systems obtained average WER of 12.39%.     

A large vocabulary system for Arabic online handwriting recognition

The success of using Hidden Markov Models (HMMs) for speech recognition application has motivated the adoption of these models for handwriting recognition especially the online handwriting that has large similarity with the speech signal as a sequential process. Some languages such as Arabic, Farsi and Urdo include large number of delayed strokes that are written above or below most letters and usually written delayed in time. These delayed strokes represent a modeling challenge for the conventional left-right HMM that is commonly used for Automatic Speech Recognition (ASR) systems. In this paper, we introduce a new approach for handling delayed strokes in Arabic online handwriting recognition using HMMs. We also show that several modeling approaches such as context based tri-grapheme models, speaker adaptive training and discriminative training that are currently used in most state-of-the-art ASR systems can provide similar performance improvement for Hand Writing Recognition (HWR) systems. Finally, we show that using a multi-pass decoder that use the computationally less expensive models in the early passes can provide an Arabic large vocabulary HWR system with practical decoding time. We evaluated the performance of our proposed Arabic HWR system using two databases of small and large lexicons. For the small lexicon data set, our system achieved competing results compared to the best reported state-of-the-art Arabic HWR systems. For the large lexicon, our system achieved promising results (accuracy and time) for a vocabulary size of 64k words with the possibility of adapting the models for specific writers to get even better results.     

An efficient speech recognition system in adverse conditions using the nonparametric regression

General Regression Neural Networks (GRNN) have been applied to phoneme identification and isolated word recognition in clean speech. In this paper, the authors extended this approach to Arabic spoken word recognition in adverse conditions. In fact, noise robustness is one of the most challenging problems in Automatic Speech Recognition (ASR) and most of the existing recognition methods, which have shown to be highly efficient under noise-free conditions, fail drastically in noisy environments. The proposed system was tested for Arabic digit recognition at different Signal-to-Noise Ratio (SNR) levels and under four noisy conditions: multispeakers babble background, car production hall (factory), military vehicle (leopard tank) and fighter jet cockpit (buccaneer) issued from NOISEX-92 database. The proposed scheme was successfully compared to the similar recognizers based on the Multilayer Perceptrons (MLP), the Elman Recurrent Neural Network (RNN) and the discrete Hidden Markov Model (HMM). The experimental results showed that the use of nonparametric regression with an appropriate smoothing factor (spread) improved the generalization power of the neural network and the global performance of the speech recognizer in noisy environments.     

面向藏语语音合成的语音基元自动标注方法

在基于隐Markov模型（Hidden Markov Model,HMM）的统计参数藏语语音合成中引入了DAEM（Deterministic Annealing EM）算法,对没有时间标注的藏语训练语音进行自动时间标注。以声母和韵母为合成基元,在声母和韵母的声学模型的训练过程中,利用DAEM算法确定HMM模型的嵌入式重估的最佳参数。训练好声学模型后,再利用强制对齐自动获得声母和韵母的时间标注。实验结果表明,该方法对声母和韵母的时间标注接近手工标注的结果。对合成的藏语语音进行主观评测表明,该方法合成的藏语语音和手工标注声、韵母时间的方法合成的藏语语音的音质接近。因此,利用该方法可以在不需要声、韵母的时间标注的情况下建立合成基元的声学模型。     

TAMEEM V1.0: speakers and text independent Arabic automatic continuous speech recognizer

This research work aims to disseminate the efforts towards developing and evaluating TAMEEM V1.0, which is a state-of-the-art pure Modern Standard Arabic (MSA), automatic, continuous, speaker independent, and text independent speech recognizer using high proportion of the spoken data of the phonetically rich and balanced MSA speech corpus. The speech corpus contains speech recordings of Arabic native speakers from 11 Arab countries representing Levant, Gulf, and Africa regions of the Arabic World, which make about 45.30 h of speech data. The recordings contain about 39.28 h of 367 sentences that are considered phonetically rich and balanced, which are used for training TAMEEM V1.0 speech recognizer, and another 6.02 h of another 48 sentences that are used for testing purposes, which are mostly text independent and foreign to the training sentences. TAMEEM V1.0 speech recognizer is developed using the Carnegie Mellon University (CMU) Sphinx 3 tools in order to evaluate the speech corpus, whereby the speech engine uses three-emitting state Continuous Density Hidden Markov Model for tri-phone based acoustic models, and the language model contains uni-grams, bi-grams, and tri-grams. Using three different testing data sets, this work obtained 7.64% average Word Error Rate (WER) for speakers dependent with text independent data set. For speakers independent with text dependent data set, this work obtained 2.22% average WER, whereas 7.82% average WER is achieved for speakers independent with text independent data set.