Improvement of phone recognition accuracy using speech mode classification

In this work, we have developed a speech mode classification model for improving the performance of phone recognition system (PRS). In this paper, we have explored vocal tract system, excitation source and prosodic features for development of speech mode classification (SMC) model. These features are extracted from voiced regions of a speech signal. In this study, conversation, extempore, and read speech are considered as three different modes of speech. The vocal tract component of speech is extracted using Mel-frequency cepstral coefficients (MFCCs). The excitation source features are captured through Mel power differences of spectrum in sub-bands (MPDSS) and residual Mel-frequency cepstral coefficients (RMFCCs) of the speech signal. The prosody information is extracted from pitch and intensity. Speech mode classification models are developed using above described features independently, and in fusion. The experiments carried out on Bengali speech corpus to analyze the accuracy of the speech mode classification model using the artificial neural network (ANN), naive Bayes, support vector machines (SVMs) and k-nearest neighbor (KNN). We proposed four classification models which are combined using maximum voting approach for optimal performance. From the results, it is observed that speech mode classification model developed using the fusion of vocal tract system, excitation source and prosodic features of speech, yields the best performance of 98%. Finally, the proposed speech mode classifier is integrated to the PRS, and the accuracy of phone recognition system is observed to be improved by 11.08%.     

Characterization and recognition of emotions from speech using excitation source information

This paper explores the excitation source features of speech production mechanism for characterizing and recognizing the emotions from speech signal. The excitation source signal is obtained from speech signal using linear prediction (LP) analysis, and it is also known as LP residual. Glottal volume velocity (GVV) signal is also used to represent excitation source, and it is derived from LP residual signal. Speech signal has high signal to noise ratio around the instants of glottal closure (GC). These instants of glottal closure are also known as epochs. In this paper, the following excitation source features are proposed for characterizing and recognizing the emotions: sequence of LP residual samples and their phase information, parameters of epochs and their dynamics at syllable and utterance levels, samples of GVV signal and its parameters. Auto-associative neural networks (AANN) and support vector machines (SVM) are used for developing the emotion recognition models. Telugu and Berlin emotion speech corpora are used to evaluate the developed models. Anger, disgust, fear, happy, neutral and sadness are the six emotions considered in this study. About 42 % to 63 % of average emotion recognition performance is observed using different excitation source features. Further, the combination of excitation source and spectral features has shown to improve the emotion recognition performance up to 84 %.     

Spotting glottal stop in Amharic in continuous speech

Glottal stop sounds in Amharic are produced due to abrupt closure of the glottis without any significant gesture in the accompanying articulatory organs in the vocal tract system. It is difficult to observe the features of the glottal stop through spectral analysis, as the spectral features emphasize mostly the features of the vocal tract system. In order to spot the glottal stop sounds in continuous speech, it is necessary to extract the features of the source of excitation also, which may require some non-spectral methods for analysis. In this paper the linear prediction (LP) residual is used as an approximation to the excitation source signal, and the excitation features are extracted from the LP residual using zero frequency filtering (ZFF). The glottal closure instants (GCIs) or epoch are identified from the ZFF signal. At each GCI, the cross-correlation coefficients of successive glottal cycles of the LP residual, the normalized jitter and the logarithm of the peak normalized excitation strength (LPNES) are calculated. Further, the parameters of Gaussian approximation models are derived from the distributions of the excitation parameters. These model parameters are used to identify the regions of the glottal stop sounds in continuous speech. For the database used in this study 92.89% of the glottal stop regions are identified correctly, with 8.50% false indications.     

Emotion recognition from speech using source, system, and prosodic features

In this work, source, system, and prosodic features of speech are explored for characterizing and classifying the underlying emotions. Different speech features contribute in different ways to express the emotions, due to their complementary nature. Linear prediction residual samples chosen around glottal closure regions, and glottal pulse parameters are used to represent excitation source information. Linear prediction cepstral coefficients extracted through simple block processing and pitch synchronous analysis represent the vocal tract information. Global and local prosodic features extracted from gross statistics and temporal dynamics of the sequence of duration, pitch, and energy values represent the prosodic information. Emotion recognition models are developed using above mentioned features separately, and in combination. Simulated Telugu emotion database (IITKGP-SESC) is used to evaluate the proposed features. The emotion recognition results obtained using IITKGP-SESC are compared with the results of internationally known Berlin emotion speech database (Emo-DB). Autoassociative neural networks, Gaussian mixture models, and support vector machines are used to develop emotion recognition systems with source, system, and prosodic features, respectively. Weighted combination of evidence has been used while combining the performance of systems developed using different features. From the results, it is observed that, each of the proposed speech features has contributed toward emotion recognition. The combination of features improved the emotion recognition performance, indicating the complementary nature of the features.     

Face Active Appearance Modeling and Speech Acoustic Information to Recover Articulation

We are interested in recovering aspects of vocal tract's geometry and dynamics from speech, a problem referred to as speech inversion. Traditional audio-only speech inversion techniques are inherently ill-posed since the same speech acoustics can be produced by multiple articulatory configurations. To alleviate the ill-posedness of the audio-only inversion process, we propose an inversion scheme which also exploits visual information from the speaker's face. The complex audiovisual-to-articulatory mapping is approximated by an adaptive piecewise linear model. Model switching is governed by a Markovian discrete process which captures articulatory dynamic information. Each constituent linear mapping is effectively estimated via canonical correlation analysis. In the described multimodal context, we investigate alternative fusion schemes which allow interaction between the audio and visual modalities at various synchronization levels. For facial analysis, we employ active appearance models (AAMs) and demonstrate fully automatic face tracking and visual feature extraction. Using the AAM features in conjunction with audio features such as Mel frequency cepstral coefficients (MFCCs) or line spectral frequencies (LSFs) leads to effective estimation of the trajectories followed by certain points of interest in the speech production system. We report experiments on the QSMT and MOCHA databases which contain audio, video, and electromagnetic articulography data recorded in parallel. The results show that exploiting both audio and visual modalities in a multistream hidden Markov model based scheme clearly improves performance relative to either audio or visual-only estimation.      

一种基于鲁棒特征的模型补偿噪声语音识别方法

针对抗噪声语音特征技术和基于MFCC特征的模型补偿技术在低信噪比时识别率不高的缺点，将抗噪声语音特征和模型补偿结合起来，提出了一种基于单边自相关序列(One—sided autocorrelation，OSA)MFCC特征的模型补偿噪声语音识别方法，以提高语音识别系统在低信噪比时的性能。对0～9十个英文数字和NOISEX92中的白噪声、F16噪声和FACTORY噪声的识别实验结果表明．本文提出的识别方法可以有效地提高OSA—MFCC识别器在噪声环境中的识别率，并且在低信噪比时其性能明显优于经过相同补偿处理的MFCC识别器。     

Articulatory-feature-based confidence measures

Confidence measures are computed to estimate the certainty that target acoustic units are spoken in specific speech segments. They are applied in tasks such as keyword verification or utterance verification. Because many of the confidence measures use the same set of models and features as in recognition, the resulting scores may not provide an independent measure of reliability. In this paper, we propose two articulatory feature (AF) based phoneme confidence measures that estimate the acoustic reliability based on the match in AF properties. While acoustic-based features, such as Mel-frequency cepstral coefficients (MFCC), are widely used in speech processing, some recent works have focus on linguistically based features, such as the articulatory features that relate directly to the human articulatory process which may better capture speech characteristics. The articulatory features can either replace or complement the acoustic-based features in speech processing. The proposed AF-based measures in this paper were evaluated, in comparison and in combination, with the HMM-based scores on phoneme and keyword verification tasks using children’s speech collected for a computer-based English pronunciation learning project. To fully evaluate their usefulness, the proposed measures and combinations were evaluated on both native and non-native data; and under field test conditions that mis-matches with the training condition. The experimental results show that under the different environments, combinations of the AF scores with the HMM-based scores outperforms HMM-based scores alone on phoneme and keyword verification.     

Speech / music classification using speech-specific features

This paper proposes the use of speech-specific features for speech / music classification. Features representing the excitation source, vocal tract system and syllabic rate of speech are explored. The normalized autocorrelation peak strength of zero frequency filtered signal, and peak-to-sidelobe ratio of the Hilbert envelope of linear prediction residual are the two source features. The log mel energy feature represents the vocal tract information. The modulation spectrum represents the slowly-varying temporal envelope corresponding to the speech syllabic rate. The novelty of the present work is in analyzing the behavior of these features for the discrimination of speech and music regions. These features are non-linearly mapped and combined to perform the classification task using a threshold based approach. Further, the performance of speech-specific features is evaluated using classifiers such as Gaussian mixture models, and support vector machines. It is observed that the performance of the speech-specific features is better compared to existing features. Additional improvement for speech / music classification is achieved when speech-specific features are combined with the existing ones, indicating different aspects of information exploited by the former.     

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%.     

Source cell phone verification from speech recordings using sparse representation

Source recording device recognition is an important emerging research field in digital media forensics. The literature has mainly focused on the source recording device identification problem, whereas few studies have focused on the source recording device verification problem. Sparse representation based classification methods have shown promise for many applications. This paper proposes a source cell phone verification scheme based on sparse representation. It can be further divided into three schemes which utilize exemplar dictionary, unsupervised learned dictionary and supervised learned dictionary respectively. Specifically, the discriminative dictionary learned by supervised learning algorithm, which considers the representational and discriminative power simultaneously compared to the unsupervised learning algorithm, is utilized to further improve the performances of verification systems based on sparse representation. Gaussian supervectors (GSVs) based on MFCCs, which have shown to be effective in capturing the intrinsic characteristics of recording devices, are utilized for constructing and learning dictionary. SCUTPHONE, which is a corpus of speech recordings from 15 cell phones, is presented. Evaluation experiments are conducted on three corpora of speech recordings from cell phones and demonstrate the effectiveness of the proposed methods for cell phone verification. In addition, the influences of number of target examples in the exemplar dictionary and size of the unsupervised learned dictionary on source cell phone verification performance are also analyzed.     

语音识别中基于发音特征的声调集成算法

提出基于发音特征的声调建模改进方法,并将其用于随机段模型的一遍解码中。根据普通话的发音特点,确定了用于区别汉语元音、辅音信息的7种发音特征,并以此为目标值利用阶层式多层感知器计算语音信号属于发音特征的35个类别后验概率,将该概率作为发音特征与传统的韵律特征一起用于声调建模。根据随机段模型的解码特点,在两层剪枝后对保留下来的路径计算其声调模型概率得分,加权后加入路径总的概率得分中。在“863-test”测试集上进行的实验结果显示,使用了新的发音特征集合中声调模型的识别精度提高了3.11%;融入声调信息后随机段模型的字错误率从13.67%下降到12.74%。表明了将声调信息应用到随机段模型的可行性。     

基于发音特征的音/视频双流语音识别模型*

构建了一种基于发音特征的音/视频双流动态贝叶斯网络(dynamic Bayesian network, DBN)语音识别模型,定义了各节点的条件概率关系,以及发音特征之间的异步约束关系,最后在音/视频连接数字语音数据库上进行了语音识别实验,并与音频单流、视频单流DBN模型比较了在不同信噪比情况下的识别效果。结果表明,在低信噪比情况下,基于发音特征的音/视频双流语音识别模型表现出最好的识别性能,而且随着噪声的增加,其识别率下降的趋势比较平缓,表明该模型对噪声具有很强的鲁棒性,更适用于低信噪比环境下的语音识别     

Speaker verification using excitation source information

In this work we develop a speaker recognition system based on the excitation source information and demonstrate its significance by comparing with the vocal tract information based system. The speaker-specific excitation information is extracted by the subsegmental, segmental and suprasegmental processing of the LP residual. The speaker-specific information from each level is modeled independently using Gaussian mixture modeling—universal background model (GMM-UBM) modeling and then combined at the score level. The significance of the proposed speaker recognition system is demonstrated by conducting speaker verification experiments on the NIST-03 database. Two different tests, namely, Clean test and Noisy test are conducted. In case of Clean test, the test speech signal is used as it is for verification. In case of Noisy test, the test speech is corrupted by factory noise (9 dB) and then used for verification. Even though for Clean test case, the proposed source based speaker recognition system still provides relatively poor performance than the vocal tract information, its performance is better for Noisy test case. Finally, for both clean and noisy cases, by providing different and robust speaker-specific evidences, the proposed system helps the vocal tract system to further improve the overall performance.     

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.     

Dictionary learning based sparse coefficients for audio classification with max and average pooling

Audio classification is an important problem in signal processing and pattern recognition with potential applications in audio retrieval, documentation and scene analysis. Common to general signal classification systems, it involves both training and classification (or testing) stages. The performance of an audio classification system, such as its complexity and classification accuracy, depends highly on the choice of the signal features and the classifiers. Several features have been widely exploited in existing methods, such as the mel-frequency cepstrum coefficients (MFCCs), line spectral frequencies (LSF) and short time energy (STM). In this paper, instead of using these well-established features, we explore the potential of sparse features, derived from the dictionary of signal atoms using sparse coding based on e.g. orthogonal matching pursuit (OMP), where the atoms are adapted directly from audio training data using the K-SVD dictionary learning algorithm. To reduce the computational complexity, we propose to perform pooling and sampling operations on the sparse coefficients. Such operations also help to maintain a unified dimension of the signal features, regardless of the various lengths of the training and testing signals. Using the popular support vector machine (SVM) as the classifier, we examine the performance of the proposed classification system for two binary classification problems, namely speech–music classification and male–female speech discrimination and a multi-class problem, speaker identification. The experimental results show that the sparse (max-pooled and average-pooled) coefficients perform better than the classical MFCCs features, in particular, for noisy audio data.     

基于发音特征的汉语声调建模方法及其在汉语语音识别中的应用

发音特征表征了语音的发音方式信息,能够辅助传统的韵律特征改善声调建模的精度。在分析汉语声韵母发音特点的基础上,将发音方式划分为19类,并提出利用阶层式多层感知器计算语音信号属于各类的后验概率,作为发音特征。之后,将发音特征与传统的韵律特征一起用于声调建模。实验结果显示,加入发音特征后,在三种不同的建模方法下声调识别的准确率提升约5%。将声调模型融入大词表连续语音识别系统后,汉字错误率有了明显的下降     

Constructing accurate and robust HMM/GMM models for an Arabic speech recognition system

Conventional Hidden Markov Model (HMM) based Automatic Speech Recognition (ASR) systems generally utilize cepstral features as acoustic observation and phonemes as basic linguistic units. Some of the most powerful features currently used in ASR systems are Mel-Frequency Cepstral Coefficients (MFCCs). Speech recognition is inherently complicated due to the variability in the speech signal which includes within- and across-speaker variability. This leads to several kinds of mismatch between acoustic features and acoustic models and hence degrades the system performance. The sensitivity of MFCCs to speech signal variability motivates many researchers to investigate the use of a new set of speech feature parameters in order to make the acoustic models more robust to this variability and thus improve the system performance. The combination of diverse acoustic feature sets has great potential to enhance the performance of ASR systems. This paper is a part of ongoing research efforts aspiring to build an accurate Arabic ASR system for teaching and learning purposes. It addresses the integration of complementary features into standard HMMs for the purpose to make them more robust and thus improve their recognition accuracies. The complementary features which have been investigated in this work are voiced formants and Pitch in combination with conventional MFCC features. A series of experimentations under various combination strategies were performed to determine which of these integrated features can significantly improve systems performance. The Cambridge HTK tools were used as a development environment of the system and experimental results showed that the error rate was successfully decreased, the achieved results seem very promising, even without using language models.     

基于修正Mel频率映射的应力影响下变异语音识别方法

文章在对应力影响下变异语音进行分频带分析的基础上,选用了可以提升语音信号中频段影响的修正Mel频率映射,并将其对应的MFCC系数作为新的语音识别特征。通过采用正常/变异语音分类器和新特征来进行变异语音识别。实验结果表明:采用修正Mel频率映射的MFCC特征改进了变异语音的识别性能。     

基于模型自适应的声效鲁棒性语音识别算法

针对声音效果变化引起的语音声学特性的改变,提出基于声学模型自适应的方法。分析了正常模式下训练的声学模型在识别其他声效模式下语音的表现;根据随机段模型的模型特性,将最大似然线性回归方法引入到随机段模型系统中,并利用自适应后的声学模型来识别对应的声效模式下的语音。在“863-test”测试集上进行的汉语连续语音识别实验显示,正常模式下训练的声学模型识别其他四种声效模式下的语音时,识别精度均有较大程度的下降;而自适应后的系统在识别对应的声效模式的语音时,识别精度有了明显的改观。表明了基于声学模型自适应的方法在解决语音识别中声音效果变化问题上的有效性。     

Spectral histogram of oriented gradients (SHOGs) for Tamil language male/female speaker classification

Gender (Male/Female) classification plays a primary vital role to develop a robust Automatic Tamil Speech Recognition (ASR) applications due to the diversity in the vocal tract of speakers. Various features including Formants (F1, F2, F3, F4), Zero Crossings, and Mel-Frequency Cepstral Coefficients (MFCCs) etc. have appeared in the literature especially for speech/signal classification/recognition. Recently Dalal et al. have proposed a feature called as Histogram of Oriented Gradients (HOG) for extracting feature from an image for efficient detection/classification of objects. We extend and apply the HOG for spectrogram of speech signal and hence called as Spectral Histogram of Oriented Gradients (SHOGs). The results of Tamil language male/female speaker classification using SHOGs features shows good improvement in the classification rate when compared to other features. The results of combination of various features with SHOGs are also promissing.