Poonlap Lamsrichan

Bio: Poonlap Lamsrichan is an academic researcher from Kasetsart University. The author has contributed to research in topics: Wavelet transform & Wavelet. The author has an hindex of 4, co-authored 14 publications receiving 271 citations.

Speech emotion recognition using convolutional long short-term memory neural network and support vector machines

Abstract: In this paper, we propose a speech emotion recognition technique using convolutional long short-term memory (LSTM) recurrent neural network (ConvLSTM-RNN) as a phoneme-based feature extractor from raw input speech signal In the proposed technique, ConvLSTM-RNN outputs phoneme- based emotion probabilities to every frame of an input utterance Then these probabilities are converted into statistical features of the input utterance and used for the input features of support vector machines (SVMs) or linear discriminant analysis (LDA) system to classify the utterance-level emotions To assess the effectiveness of the proposed technique, we conducted experiments in the classification of four emotions (anger, happiness, sadness, and neutral) on IEMOCAP database The result showed that the proposed technique with either of SVM or LDA classifier outperforms the conventional ConvLSTM-based one

A study of support vector machines for emotional speech recognition

Abstract: In this paper, efficiency comparison of Support Vector Machines (SVM) and Binary Support Vector Machines (BSVM) techniques in utterance-based emotion recognition is studied. Acoustic features including energy, Mel-frequency cepstral coefficients (MFCC), Perceptual linear predictive (PLP), Filter bank (FBANK), pitch, their first and second derivatives are used as frame-based features. Four basic emotions including anger, happiness, neutral and sadness in Interactive Emotional Dyadic Motion Capture (IEMOCAP) database are selected for training and evaluating in our experiments. The best accuracy of emotional speech recognition is 58.40% in average from SVM with polynomial kernel. Energy features combination with FBANK, pitch and their first and second derivatives features are the most suitable for computing utterance feature. Binary Support Vector Machines (BSVM) techniques show accuracy improvement in some emotions, such as sadness and happiness emotion.

A bi-lingual Thai-English TTS system on Android mobile devices

Abstract: This paper presents a bi-lingual Thai-English text-to-speech synthesis (TTS) system on Android mobile devices. The system deploys a Thai text processor and a well-known open-source English text processor, which can analyzes English text at high intelligibility. With hidden Markov model (HMM) based speech unit and audio streaming optimization, it can synthesize highly smoothed sounds at a fast response. This paper reveals the optimization of important components. Conditional random fields (CRF) successfully used in Thai word segmentation and a syllable-pattern based statistical modeling for Thai grapheme-to-phoneme conversion are assessed. Several types of speech parameters are compared for best performance. The optimized system produced as high as 3.68 mean opinion score (MOS) with response less than 2 seconds on both high and low specification devices.

Energy-Efficient Data Compression in Clustered Wireless Sensor Networks using Adaptive Arithmetic Coding with Low Updating Cost

Abstract: This work presents a study of energy reduction technique using data compression based on arithmetic coding in clustered wireless sensor networks to maximize the network's lifetime. Initially, a simulation approach is used to investigate the effect of multiple data types found in environmental monitoring application on data compression and the effect of cluster's parameters on their energy consumption. This study points out the important of probability models of multiple sensor data such as temperature and relative humidity on the arithmetic coding's performance. The investigation results provide insights for designing an energy-efficient arithmetic coding framework that is suitable for compressing multiple data types in clustered multi-hop wireless sensor networks. Finally,an implementation of an adaptive local data compression algorithm derived from our findings and design framework on a set of four TinyOS based Tmote Sky wireless sensor nodes equipped with temperature and relative humidity sensors is presented with approximately 54 percent data compression results.

Big Data Meet Green Challenges: Greening Big Data

Abstract: Nowadays, there are two significant tendencies, how to process the enormous amount of data, big data, and how to deal with the green issues related to sustainability and environmental concerns. An interesting question is whether there are inherent correlations between the two tendencies in general. To answer this question, this paper firstly makes a comprehensive literature survey on how to green big data systems in terms of the whole life cycle of big data processing, and then this paper studies the relevance between big data and green metrics and proposes two new metrics, effective energy efficiency and effective resource efficiency in order to bring new views and potentials of green metrics for the future times of big data.

Wireless sensor network for structural health monitoring: A contemporary review of technologies, challenges, and future direction:

Abstract: The importance of wireless sensor networks in structural health monitoring is unceasingly growing, because of the increasing demand for both safety and security in the cities. The speedy growth of ...

Compression in wireless sensor networks: A survey and comparative evaluation

Abstract: Wireless sensor networks (WSNs) are highly resource constrained in terms of power supply, memory capacity, communication bandwidth, and processor performance. Compression of sampling, sensor data, and communications can significantly improve the efficiency of utilization of three of these resources, namely, power supply, memory and bandwidth. Recently, there have been a large number of proposals describing compression algorithms for WSNs. These proposals are diverse and involve different compression approaches. It is high time that these individual efforts are put into perspective and a more holistic view taken. In this article, we take a step in that direction by presenting a survey of the literature in the area of compression and compression frameworks in WSNs. A comparative study of the various approaches is also provided. In addition, open research issues, challenges and future research directions are highlighted.