Recognition of intonation patterns in Thai utterance
نویسندگان
چکیده
Thai intonation can be categorized as paralinguistic information of F0 contour of the utterance. There are three classes of intonation pattern in Thai, the Fall Class, the Rise Class, and the Convolution Class. This paper presents a method of intonation pattern recognition of Thai utterance. Two intonation feature contours, extracted from F0 contour, were proposed. The feature contours were converted to feature vector to use as input of neural network recognizer. The recognition results show that an average recognition rate is 63.4% for male speakers and 75.4% for female speakers. The recognizer can recognize the Fall Class from the others better than distinguish between the Rise Class and the Convolution Class.
منابع مشابه
Disambiguation of Korean utterances using automatic intonation recognition
The paper describes a research on a use of intonation for disambiguating utterance types of Korean spoken sentences. Based on tilt intonation theory [8], two related but separate experiments were performed, both using the Hidden Markov Model training technique. In the first experiment, a system is established so that rough boundary positions of major intonation events are detected. Subsequently...
متن کاملOn utterance-final intonation in tonal and non-tonal dia-lects of Kammu
In this study we investigate utterance-final intonation in two dialects of Kammu, one tonal and one non-tonal. While the general patterns of utterance-final intonation are similar between the dialects, we do find clear evidence that the lexical tones of the tonal dialect restrict the pitch range and the realization of focus. Speaker engagement can have a strong effect on the utterance-final acc...
متن کاملEstimating speaker-specific intonation patterns using the linear alignment model
Modeling speaker-specific intonation is important in several areas, including speaker identification, verification, and imitation using text-to-speech synthesis. However the choice of the intonation model and the estimation of its parameters from spontaneous speech remains a challenge. We propose a way to estimate speaker-specific intonation parameters for a particular superpositional model, th...
متن کاملTone recognition in Thai continuous speech based on coarticulaion, intonation and stress effects
Tone recognition is a critical component for speech recognition in a tone language. One of the main problems of tone recognition in continuous speech is that several interacting factors affect F0 realization of tones. In this paper, we focus on the coarticulatory, intonation, and stress effects. These effects are compensated by the tone information of neighboring syllables, the adjustment of F0...
متن کاملRecognizing the Emotional State Changes in Human Utterance by a Learning Statistical Method based on Gaussian Mixture Model
Speech is one of the most opulent and instant methods to express emotional characteristics of human beings, which conveys the cognitive and semantic concepts among humans. In this study, a statistical-based method for emotional recognition of speech signals is proposed, and a learning approach is introduced, which is based on the statistical model to classify internal feelings of the utterance....
متن کامل