Abstract

In this paper, we investigate two different recurrent neural network (RNN) architectures: Elman RNN and recently proposed clockwork RNN [1] for statistical parametric speech synthesis (SPSS). Of late, deep neural networks are being used for SPSS which involve predicting every frame independent of the previous predictions, and hence requires post-processing for ensuring smooth evolution of speech parameters. RNNs, on the other hand, are intuitively better suited for the task as they inherently model temporal dependencies, but were restricted in use because of the difficulty in training. Lately, techniques such as sparse initialization, Nesterov’s accelerated gradient, gradient clipping and leaky integration (LI) have been shown to overcome this difficulty. We study the utility of these techniques for SPSS task. In addition, we show that clockwork RNN is equivalent to an Elman RNN with a particular form of LI. This perspective enables us to understand the reason why a simple Elman RNN with LI units performs well on sequential tasks.

Abstract

During production of emotional speech there are deviations in the components of speech production mechanism when compared to normal speech. The objective of this study is to capture the deviations in features related to the excitation source component of speech, and to develop a system for automatic recognition of emotions based on these deviations. The emotions considered for this study are: anger, happy, neutral and sad. The study shows that there are useful features in the deviations of the excitation source features at subsegmental level, and they can be exploited to develop an emotion recognition system. A hierarchical binary decision tree approach is used for classification.

Abstract

Plug loads accounts for 20% to 30% of building energy consumption and has an increasing trend. Automatic plug load identification is one of the technique for effectively managing the plug load consumption.There are several studies on Non Intrusive load monitoring (NILM) but limited studies on Intrusive load monitoring (ILM). ILM is a technique that uses a low end power meter on every plug load to monitor it’s power consumption. In this paper, machine-l earning techniques are applied on low-frequency ILM data for identification of plug load and it’s state (ON/Sleep).The results show the identification accuracies are close to 98%

Abstract

Measuring “similarity” has been established as a fundamental problem and has been widely studied. In this paper we propose a novel approach for establishing similarity in context of citation network. With the rapidly growing size of academic literature, the problem of finding similar research papers has become a challenging task. Research papers in a citation network often form communities based on an under lying concept. Our research shows that dispersion based similarity measure can be used as a strong measure for finding similar papers based on similar connectivity in those communities and structural relevance of the citation network. Our results show that our approach works better than other conventional link-based similarity measures both quantitatively and qualitatively.One of the direct benefits of this research is to support the highly specialised information needs of a scholarly researcher working in a specialised field of research.

Abstract

Energy utilization can be improved by precise plug load monitoring and control. Plug load energy consumption is nearly 30% of the total building energy consumption. Therefore, plug load identification is a key requirement for energy conservation in buildings. Intrusive load monitoring techniques identify loads precisely but have not been tested widely so far for their performance in changing operating conditions. Hence, the present research proposes a robust low frequency intrusive load monitoring technique to identify load accurately. A smart power strip using proposed load identification technique is designed and developed. Linear regression is applied on the acquired data to capture the behavioral trends of a particular device more explicitly and concisely. Further, weighted K-NN classifier is applied on the transformed data set for device. Experimental results show that the proposed algorithm performs better than the standard classifiers, and can offer tangible savings.

Abstract

A high degree of robustness is a prerequisite to operate speech and language processing systems in practical environments. Performance of these systems is highly influenced by varying and mixed background environments. In this pa-per, we put forward a robust method for automatic language identification in various background environments. Combined temporal and spectral processing method is used as a preprocessing technique for enhancing the degraded speech. Language discriminative information in high sonority regions of speech is used for the task of language identification. Sonority regions are regions of speech whose signal energy is high and these regions are less influenced by background environments.Spectral energy of formants in the glottal closure regions is employed as an acoustic correlate for the detection of sonority regions of speech. In this paper performance of the LID system is studied in various background environments like clean room, car factory,high frequency,pink and white noise environments. In this work, Indian Institute of Technology Kharagpur - Multi LingualI ndian Language Speech Corpus (IITKGP-MLILSC) is used for building language identification system. Noise speech samples from the NOISEX database are employed in the present study.The performance of the proposed method is quite satisfactory compared to existing approaches.

Abstract

Most of the Indian languages are originated from Devanagari, the script of the Sanskrit language. In-spite of similarity in phoneme sets, every language its own influence on the phonotactic constraints of speech in that language. A modelling technique that is capable of capturing the slightest variations imparted by the language is a prerequisite for developing a language identification system (LID). Use of Gaussian mixture modelling technique with a large number of mixture components demands a large training data for each language class, which is hard to collect and handle. In this work, phonotactic variations imparted by the different languages are modelled using Gaussian mixture modelling with a universal background model (GMM-UBM) technique.In GMM-UBM based modelling certain amount of data from all the language classes is pooled to develop a universal background model (UBM) and the model is adapted to each class. Spectral features (MFCC) are employed to represent the language specific phonotactic information of speech in different languages. During the present study, LID systems are developed using the speech samples from IITKGP-MLILSC. In this work, performance of the proposed GMM-UBM based LID system is compared with conventional GMM based LID system. An average improvement of 7–8% is observed due to the use of UBM-based modelling of developing a LID system.

Abstract

In recent past a lot of scientific attention is paid on recognizing the emotional state of the speaker from his speech. Emotion recognition is a challenging task as human emotions are complex, subtle and emotive state in human speech does not persist long. So it is important to study the presence of emotion identifiable information in smaller segments of speech. This study is aimed at studying the presence of emotional specific information with relevance to the position of the word in the utterance. During the present study, spectral features are employed to represent emotion specific information in speech. Spectral features from smaller speech segments of speech based on their position in the utterance are employed to study the presence of emotion in speech. Due to the lack of adequate data in small speech segments to support conventional GMM during the course of present study Gaussian mixture modeling with a universal background model (GMM-UBM) is used for developing a emotion recognition system. Speech data from IITKGP-SESC is used during the course of the present study. During the present study 4 (Anger, Fear, Happy and Neutral) emotions are considered.

Abstract

Emotions in human speech are short lived. In an emotive utterance, the emotive gestures produced due to the emotive state of the speaker persists only to a shorter duration. In this study, the regions of an utterance that are highly influenced by the emotive state of the speaker are detected. These regions are labeled as emotionally significant regions. Data from the detected emotionally significant regions is used for developing an emotion recognition system. Physiological constraints of human speech production system are explored for detecting the emotionally significant regions of an utterance. Spectral features from the detected emotionally significant regions are used to develop an emotion recognition system. A significant improvement in the performance of the emotion recognition system is observed inthe present approach. An average improvement of 11% is in noted owing to the use of data from emotionally significant regions while developing an emotion recognition system. Gaussian mixture modelling (GMM) technique is employed to develop an emotion recognition system. During the present study, speech samples from Berlin emotion speech database (EMO-DB) are used. Four basic emotions such as anger, happy, neutral and fear are considered for study

Abstract

In this paper we propose a technique for spectral envelope estimation using maximum values in the sub-bands of Fourier magnitude spectrum (MSASB). Most other methods in the literature parametrize spectral envelope in cepstral domain such as Mel-generalized cepstrum etc. Such cepstral domain representations, although compact, are not readily interpretable. This difficulty is overcome by our method which parametrizes in the spectral domain itself. In our experiments, spectral envelope estimated using MSASB method was incorporated in the STRAIGHT vocoder. Both objective and subjective results of analysis-by-synthesis indicate that the proposed method is comparable to STRAIGHT. We also evaluate the effectiveness of the proposed parametrization in a statistical parametric speech synthesis framework using deep neural networks.