Montri Karnjanadecha

Improvement of Thai speech emotion recognition by using face feature analysis

The fact that in Thai language emotions are not usually manifested, mostly because any emotion would interfere with meaning otherwise, makes this language very difficult for any kind of emotion recognition. Our proposed Thai emotion recognition system consists of two parts - speech emotion recognition and improvements of the system using face feature analysis. For this purpose audiovisual Thai emotion database was recorded. Speech emotion recognition is based on calculating fundamental frequency, zero crossing rate and energy from short-time wavelet signals, and shows great system results with accuracy of 97.8%. Our current research activities are directed to improving the accuracy of the overall system using face feature analysis, therefore showing that vision is as crucial as hearing is for expressing and recognizing any emotion.

Dual transmission model and related spectral content of the fetal heart sounds

A dual transmission model of the fetal heart sounds is presented in which the properties of the signals received on a sensor, installed on the maternal abdominal surface, depend upon the position of the fetus. For a fetus in the occiput anterior position, the predominant spectral content lies in the frequency band 16-50Hz (impact mode), but for a fetus in the occiput posterior position, it lies in the frequency band 80-110Hz (acoustic mode). Signal processing comprises digital bandpass filtering, matched filtering, Teager energy operator, autocorrelation, and figure of merit algorithms. The digital filter permits the user to select the frequency band that best conforms to the prevailing signal mode. Clinical tests on twelve patients, with some in the occiput anterior and some in the occiput posterior fetal positions, support the validity of the dual transmission model.

Classification of Fresh Aromatic Coconuts by Using Polynomial Regression

This paper present the classification of fresh aromatic coconuts into 3 types: single layer, double layer and one and a half layer by inspecting colors at the bottom of coconuts. We take the photos the bottom of coconuts in RGB mode, change the colors into the HSV mode, and then place 4 circles into the image. The 20 photos of each type are used to generate the relation of the rings for each type by using polynomial regression. Finally, we use the polynomial equations to test new 100 fresh aromatic coconuts, the result is 11.76% errors for single layer, 18.6% for one and a half layer and 18.18% error for double layers.

Acoustic Features for Music Emotion Recognition and System Building

We are faced with a massive growth of musical data in the form of digital files. Accurate metadata labeling of music archives is necessary in order to make digital music searchable and to be efficiently organized, not only by file name or song title but in deeper detail, such as genre, artist, and types of instrument. Music related emotional terms are frequently used as search keywords, especially for non-vocal music such as film soundtracks. Music emotion recognition is a challenging issue because there are no standard categories, and how music evokes emotion is poorly understood, and related to many factors. Manually labeling music emotion by an expert is infeasible when dealing with huge amounts of music and when opinion on the same musical piece can vary from person to person. A system that is able to reveal the hidden relationships between music-related factors and emotional perception is required. This paper discuses background knowledge on psychology, musicology, and acoustic feature extraction by MATLAB toolboxes. It also includes construction guidelines for a music emotion classification system.

Constructing a phonetic transcribed text corpus for Southern Thai dialect Speech Recognition

This paper presents the progress of the development of a Southern Thai dialect speech corpus to build automatic speech recognition. This is the recent Southern Thai dialect pronunciation dictionary containing more than 15,000 words, including words from a Southern Thai dialect dictionary, academic thesis papers, and online article resources. In this study, a hybrid technique was proposed to construct a phonetic of a Southern Thai dialect in building a Southern Thai Dialect Continuous Speech Recognition corpus. This system achieves 97.35% word accuracy in 6,500 sentences of the Southern Thai dialect. The complete resources are freely used available for the purpose of any research in order to encourage speech technology research in Southern Thai dialect.

An Angle Measurement of Hard Disk's Head in the HSA Using Image Processing

Hard disks head (HDH) is one of the most important parts in a hard disk. Because the floating height between the HDH and the platter is very narrow, it is necessary to make sure that the HDHs pitch angle and roll angles are from -0.05˚ to 0.05˚ on the Head Stack Assembly (HSA). The maximum error is 0.0226˚ for pitch angle and 0.0288˚ for roll angle according to our experimental results. This paper presents an image processing techniques for measuring the pitch and roll angle of the HDH. Keywords - hard disks head, angle measurement, image processing.

A low-cost solution of motor control system for Robocup robots

This paper describes a motor control system for Robocup (small-size league) robot. Our aim was to build high performance robots at a minimal cost. We achieved a low-cost design with the use of cheap motors and use of small DC motors for speed feedback. All controls were completed by a single microcontroller. We also established predictive controllers in software to improve the systempsilas performance.