Visual isolated word recognition. (2009)
Foo, Say Wei., Lim, Chee Chuan.
In this paper, the performance of neural network for isolated word recognition based on the movement of the mouth is investigated. Ten isolated words "zero" to "nine" form the vocabulary. The three...
A simplified viterbi matching algorithm for word partition in visual speech processing. (2009)
Foo, Say Wei., Yong, Lian., Liang, Dong.
In this paper, a novel simplified Viterbi matching algorithm for sequence partition is presented. This method connects the states of different HMMs to model the observed sequence. The method is...
Non-blind audio-watermarking using compression-expansion of signals. (2008)
One of the advantages of non-blind watermarking scheme is that the original signal must be available to extract the watermark. As such, the possession of the original signal becomes important in the...
Audio-watermarking with stereo signals. (2008)
In this paper, we propose a non-blind audiowatermarking scheme based on compression-expansion of stereo signals. We also demonstrate the general methods for embedding binary code in single channel...
Distraction in GVF-based segmentation. (2007)
Cheng, Jierong., Foo, Say Wei.
For complex images with the presence of stronger edges near the desired boundary, the snake used in gradient vector flow (GVF)-based segmentation method may be distracted to converge wrongly on the...
Two-channel time-frequency audio watermarking. (2007)
Hertanto, Richard Nathaniel., Foo, Say Wei.
The time-frequency audio watermarking scheme [1] operates by removing and adding samples to the frequency spectrum of the audio signals (compression-expansion) as a way to embed watermarks. The...
Dynamic directional gradient vector flow for snakes. (2006)
Cheng, Jierong., Foo, Say Wei.
Snakes, or active contour models, have been widely used in image segmentation. However, most present snake models do not discern between positive and negative step edges. In this paper, a new type of...
Cheng, Jierong., Foo, Say Wei., Krishnan, Shankar M.
In this paper,an automated method of boundary detection of the left ventricle (LV) is proposed. The method uses a watershed transform and morphological operation to locate the region containing the...
Markovian level set for echocardiographic image segmentation. (2006)
Cheng, Jierong., Foo, Say Wei.
Owing to the large amount of speckle noise and ill-defined edges present in echocardiographic images, computerbased boundary detection of the left ventricle (LV) has proved to be a challenging...
Markovian level set : a new method for boundary detection from echocardiographic images. (2006)
Cheng, Jierong., Foo, Say Wei.
Computer-based boundary detection in Echocardiographic Images is a challenging problem, due to the large amount of noise and the poor contrast presented. In this paper, a Markovian level set method...
Recognition of musical instruments. (2006)
Harya, Wicaksana., Septian, Hartono., Foo, Say Wei.
In this paper an automated method to recognize the musical instruments playing the musical signals is presented. Various features of the musical instruments and musical signals are investigated. The...
A Two-Channel Training Algorithm for Hidden Markov Model and Its Application to Lip Reading (2005)
Liang Dong, Say Wei Foo, Yong Lian
Hidden Markov model (HMM) has been a popular mathematical approach for sequence classification such as speech recognition since 1980s. In this paper, a novel two-channel training strategy is proposed...
Say Wei Foo, Yong Lian, Liang Dong
Automatic speech recognition (ASR) is a machine learning problem. Hidden Markov models are widely used in ASR. Visual speech (lip reading) can be used to improve robustness of speech recognizers....
A blind audio watermarking scheme using peak point extraction. (2005)
Foo, Say Wei., Xue, Feng., Li, Mengyuan.
In this paper, a blind digital audio watermarking scheme is proposed. A psychoacoustic model is used to ensure the imperceptibility of the embedded watermark. The watermark is embedded in the...
Active noise cancellation headset. (2005)
Foo, Say Wei., Senthilkumar, T. N., Averty, Charles.
In this paper, a new design for headset with active noise cancellation capability is presented. For the proposed headset, a Variable-Step-Size Normalized Least-Mean-Square (VSS-NLMS) algorithm is...
Cheng, Jierong., Foo, Say Wei., Krishnan, Shankar M.
In the detection of the boundary of the left ventricle from echocardiographic images, the crucial step is to determine the region of interest (ROI) or the center point (CP) of the left ventricle. In...
A Two-Channel Training Algorithm for Hidden Markov Model and Its Application to Lip Reading (2005)
Liang Dong, Say Wei Foo, Yong Lian
Hidden Markov model (HMM) has been a popular mathematical approach for sequence classification such as speech recognition since 1980s. In this paper, a novel two-channel training strategy is proposed...
Recognition of visual speech elements using adaptively boosted hidden Markov models. (2004)
Foo, Say Wei., Lian, Yong., Dong, Liang.
The performance of automatic speech recognition (ASR) system can be significantly enhanced with additional information from visual speech elements such as the movement of lips, tongue, and teeth,...
Echocardiographic image sequence segmentation by multiscale directional snake. (2004)
Cheng, Jierong., Foo, Say Wei., Krishnan, Shankar M.
In this paper, a new algorithm named as ‘multiscale directional snake’ is proposed for segmentation of chocardiographic images by computer. The method makes use of morphological operations to...
Directional gradient vector flow for snakes. (2004)
Cheng, Jierong., Foo, Say Wei., Krishnan, Shankar M.
Snakes, or active contour models, have been widely used in image segmentation. However, most present snake models cannot discern between positive and negative step edges. In this paper, a new type of...
Automated breast masses segmentation in digitized mammograms. (2004)
Zhang, Han., Foo, Say Wei., Krishnan, S. M., Thng, Choon Hua.
In this paper, an automated segmentation method is proposed. The method is applied to the segmentation of breast masses in digitized mammograms and it operates on the whole mammograms instead of...
Audio watermarking using time frequency compression expansion. (2004)
Foo, Say Wei., Ho, Shuet Mun., Ng, Ling Mei.
A novel non-blind digital audio watermarking scheme using time-frequency based operations is presented. The digital audio samples are first segmented into frames and the frames are transformed into...
Boundary detection in echocardiographic images by directional gradient vector flow. (2004)
Cheng, Jierong., Foo, Say Wei., Krishnan, Shankar M.
Snakes, or active contour models, have been widely used in image segmentation. In this paper, a new type of dynamic external force for snakes named directional gradient vector flow (DGVF) is proposed...
Recognition of piano notes with the aid of FRM filters. (2004)
Foo, Say Wei., Lee, Edwin Wei Thai.
An approach to recognition of piano notes using the Analysis-by-Synthesis technique is described in this paper. It is assumed that several notes may be played simultaneously and only a single signal...
Speech recognition using adaptively boosted classifier. (2003)
In this paper, a novel approach for speaker recognition is proposed. The system makes use of adaptive boosting (AdaBoost)and multilayer perceptions (MLP) as classifier for closed set, text-dependent...
A boosted multi-HMM classifier for recognition of visual speech elements. (2003)
A novel boosted classifier using multiple Hidden Markov Models (HMMs) is reported in this paper. The composite HMMs are specially trained to highlight certain group of training samples with the...
Nwe, Tin Lay., Foo, Say Wei., De Silva, Liyanage C.
In this paper, a novel system for detection of human stress and emotion in speech is proposed. The system makes use of FFT based linear short time Log Frequency Power Coefficients (LFPC) and TEO...
A bilingual speech recognition system for English and Tamil. (2003)
Kumar, Santosh C., Foo, Say Wei.
This paper describes the details of a bilingual speech recognition system, AmritaRec, developed for English and Tamil. The performance results of the system is compared with that of a monolingual...
A two-channel training algorithm for hidden Markov model to identify visual speech elements. (2003)
Foo, Say Wei., Yong, Lian., Dong, Liang.
A novel two-channel algorithm is proposed in this paper for discriminative training of Hidden Markov Models (HMMs). It adjusts the symbol emission coefficients of an existing HMM to maximize the...
Classification of stress in speech using linear and nonlinear features. (2003)
Nwe, Tin Lay., Foo, Say Wei., De Silva, Liyanage C.
In this paper, three systems for classification of stress in speech are proposed. The first system makes use of linear short time Log Frequency Power Coefficients (LFPC), the second employs Teager...
Cross-speaker viseme mapping using hidden Markov models. (2003)
Dong, Liang., Foo, Say Wei., Yong, Lian.
In this paper, a method of mapping visual speech between different speakers is proposed. This approach adopts Hidden Markov Model (HMM) to model the basic visual speech element – viseme. Some...
Modeling continuous visual speech using boosted viseme models. (2003)
Dong, Liang., Foo, Say Wei., Yong, Lian.
In this paper, a novel connected-viseme approach for modeling continuous visual speech is presented. The approach adopts AdaBoost-HMMs as the viseme models. Continuous visual speech is modeled by...
Using 3D deformable template trellis to describe the movement of the lip. (2003)
Foo, Say Wei., Yong, Lian., Liang, Dong.
Lip reading becomes a fascinating research area in multimedia since 1990’s. It is proven with many experiments that even small effort is made toward the incorporation of visual signal, the...
Speaker recognition using adaptively boosted decision tree classifier. (2002)
In this paper, a novel approach for speaker recognition is proposed. The approach makes use of adaptive boosting (AdaBoost) and C4.5 decision trees for closed set, text-dependent speaker recognition....
Transcription of polyphonic signals using fast filter bank. (2002)
Foo, Say Wei., Lee, Edwin Wei Thai.
In this paper, a novel approach to the transcription of polyphonic signals is investigated. The method makes use of top-down analysis and bottom-up reconstruction of the time-frequency contents of...
In this paper, a novel two-channel learning method for hidden Markov model (HMM) is proposed. This method is specially designed to train HMMs for fine recognition from similar observations. The...
Characterization with embedded synchronization for read–write channels. (2001)
Lin, Maria Yu., Bergmans, J. W. M., Mathew, George., Foo, Say Wei.
As densities and data rates increase, accurate characterization of the read/write process becomes increasingly critical in magnetic/optical recording. In this paper, we propose an advanced...
Lim, Yong Ching., Yu, Ya Jun., Zheng, H. Q., Foo, Say Wei.
The effective length of a filter designed using the frequency-response masking technique is very high and requires a very large number of delay elements. In this paper, we present some useful...
Separation of periodic signals using non-uniform embedding. (2001)
In this paper, a new method to separate overlapped periodic signals is presented. The method makes use of different embedding dimensions to describe different periodic signals and a feedback...
Recovery of speech signal from contaminated input. (2001)
Lee, Huan Hong., Foo, Say Wei.
In this paper, investigation is made on the application of an adaptive FIR filter to estimate the desired audio signal, in this case speech signal, from an input of the signal contaminated with...
Speech based emotion classification. (2001)
Tin, Lay Nwe., Foo, Say Wei., De Silva, Liyanage C.
In this paper, a speech based emotion classification method is presented. Six basic human emotions including anger, dislike, fear, happiness, sadness and surprise are investigated. The recognizer...
An adaptive audio watermarking system. (2001)
Foo, Say Wei., Yeo, Theng Hee., Huang, Dong Yan.
An adaptive and content-based audio watermarking system based on echo hiding is presented. The proposed system aims to overcome the problems of audible echoes in a simple echo hiding approach....
An innovative approach to transcription of polyphonic signals. (2001)
In this paper, a new method is proposed to separate and identify polyphonic signals generated by an acoustic piano. The method consists of a top down extraction architecture using a...
Compensation of MR head non-linearities using a saturable transfer function. (2000)
Lin, Maria Yu., Foo, Say Wei., Mathew, George., Bergmans, J. W. M., Ye, Weichun.
It is well known that the nonlinear behavior of Magneto-Resistive (MR) heads result in the isolated pulses having asymmetry and saturation. These nonlinear effects can be quantified by measuring the...
Software risk assessment model. (2000)
Foo, Say Wei., Arumugam Muruganantham.
The software industry has grown rapidly with the increasing applications of information technology. Just like any development projects. Failure to develop software according to specifications or...
Multiple regression models for electronic product success prediction. (2000)
Lo, Frank Cheong Wah., Foo, Say Wei., Bauly, John A.
As the cost of failure in new product development is very high, product developers are looking for good product success/failure prediction models. The general direction of search is towards Knowledge...
Typical management shortfalls in New Product Development (NPD) and their avoidance. (2000)
This paper briefly reviews one model of the total NPD process, viewed as a "strategic business process". The model is based upon that used by the UK Time to Market Association, [1][2], and its member...
Measurement of MR induced nonlinearities using a novel single neuron model. (2000)
Lin, Maria Yu., Foo, Say Wei., Ye, Weichun., Mathew, George., Bergmans, J. W. M.
The MR heads induced nonlinear effects include asymmetry and saturation. Superposition model with nonlinear transfer function are used to describe the magnetic recording channel with MR induced...
Hybrid method of tolerance design. (1999)
A new method of tolerance design, which we call the Hybrid method that accelerates the attainment of optimal design center, is proposed. The method combines geometry search with statistical process....
Computer-aided design of a mass production circuit. (1999)
Tolerance of the circuit components can sometimes bring about costly consequences in mass production. To tackle this problem, recent efforts have been dedicated to develop tolerant circuit design in...
Recognition of piano notes. (1999)
Foo, Say Wei., Wong, Pui Leng.
In this paper, an algorithm for the recognition of musical sound from piano is proposed. The music signal is first digitally sampled and transformed from the time domain to the frequency domain using...
Optimum quantizer design for pre-echo control in a low bit rate audio coding system. (1999)
Lim, Ai Chye., Huang, Dong Yan., Foo, Say Wei.
In a low bit-rate audio transform coding system, the quantization noise is spread out over the entire analysis block in the time domain after the inverse transformation in the decoder. This increases...
Musical sound recognition. (1999)
In this paper, two different methods to implement one particular aspect of music transcription – musical note recognition – are presented. Both methods make use of existing algorithms in the form...
An efficient method for parametric yield gradient estimation. (1999)
A novel method to improve the yield gradient estimation in parametric yield optimization is proposed. By introducing some deterministic information into the conventional Monte Carlo method and fully...
HMM speech recognition with reduced training. (1997)
One of the problems faced in automatic speech recognition is the amount of training required to adapt the machine to the speaker way of pronunciation. To a certain extent, the accuracy of correct...