Search

Mood disorders include unipolar disorder (UD) and bipolar disorder (BD). In this work, database collection of facial expressions and speech responses based on eliciting emotional videos for mood disorder detection is proposed. First, the chi-squared test was used to evaluate the agreement among manual selection of eliciting emotional videos from 11 candidate emotion videos. Six eliciting emotional videos were selected and then applied to elicit speech responses and facial expressions of the patients. In this paper, the collection procedure of the elicited data is described and the statistics of the database is provided. The collected mood database can be used in mood disorder detection and studies on emotions from elicitation.

Early detection of changes in patient's condition and making an accurate nursing diagnosis are crucial in clinical practice. Current protocol in nursing diagnosis highly depends on caregiver's experience. Demand on processing a large number of nursing documents and transferring expertise has become an important issue. This paper applies computational linguistics approach to extract knowledge and structural model of diagnosis for assisted assessment and change detection of affective expression. A corpus with peer-reviewed case reports and nursing documents were collected for analysis. The evaluation results show that narrative texts with inherent structure and emotional information can be systematically extracted.

Monitoring health conditions and events of grandparent-headed family is important to increase their quality of life and reduce care burdens. Affective episodes are significant indexes in monitoring behavior changes. In this paper, we propose an information retrieval approach to extract affect words from speech and written text to provide quantitative evidence of physical functions and social interactivity for living support and the health related quality of life assessment. Hidden Markov model with a developed behavior grammar network was adopted to transcribe speech. Combined with written texts, an adjusted term-frequency and a sliding window method were performed to extract and quantify affect words. A quantitative index scored by trigger pair approach was applied to assess affective episodes with time and place. Experimental results and case study revealed that the proposed approach shows encouraging potential in monitoring daily activity and family dialog. Its extension may provide an alternative way to obtain implicit information of emotional expression between a family.

This paper presents an approach to automatic prediction of the traits the listeners attribute to a speaker they never heard before. In previous research, the Big Five Inventory (BFI), one of the most widely used questionnaires, is adopted for personality assessment. Based on the BFI, in this study, an artificial neural network (ANN) is adopted to project the input speech segment to the BFI space based on acoustic-prosodic features. Personality trait is then predicted by estimating the BFI scores obtained from the ANN. For performance evaluation, the BFI with two versions (one is a complete questionnaire and the other is a simplified version) were adopted. The experiments were performed over a corpus of 535 speech samples assessed in terms of personality traits by experienced subjects. The results show that the proposed method for predicting the trait is efficient and effective and the prediction accuracy rate can achieve 70%.

Based on the clinic data, 69 % - 80 % stroke patients in Taiwan have the muscle weakness symptom of upper limb. Stroke patients are able to restore their independent living if they keep formal rehabilitation without interruption. However, due to the tedious and painful process, stroke patients may pause or suspend their rehabilitation therapy. Moreover, physiatrists in Taiwan hospitals must take care of more than one patient at the same time. This paper is motivated to propose a motor assessment system for stroke upper-limb rehabilitation. Our implementation is integrated with the Complete Minnesota Dexterity Test to quantify patients' recovery status automatically. Therefore, our implementation is able to reduce the workload of physiatrists effectively. In addition to assessment scale score, the proposed system is able to record more details, e.g., the tremor status and the grip strength, for doctors to evaluate the following rehabilitation therapy. Finally, a body-sensed game is provided to increase interest to the rehabilitation process.

This study presents a fast speaker clustering method based on multidimensional scaling. Speech segments are trained as initial acoustic models. MDS is utilized to transform acoustic models to a space with the coordinate best preserve the distances or dissimilarity between models. Speaker clusters are clustered using vector quantization on the MDS coordinates and the acoustic speaker models are trained on MFCCs features for each cluster. Experimental results show the proposed method outperforms the baseline speaker clustering method in lower execution time.

In Taiwan, speech technology development has been focused on Mandarin system; there is lack of related researches of domestic Taiwanese speech. This study is aimed to develop a HMM-based Taiwanese speech synthesis system, and the established sub-syllables units' database is used to be the synthesis units for all of the possible Taiwanese semantic syllables. The multiple accent corpus-based databases were developed by all combination of basic phonemes of vowels, consonants and 8-tones in Modern Literal Taiwanese (MLT) in this study. Based on the phoneme tables, this research provide fundamental database for speech analysis and synthesis in Taiwanese spelling systems. The principles and procedures in constructing phonetically balanced sentences are constituted by Taiwanese grammatical in syntax and semantics, and employ all the possible Taiwanese. Pattern recognition was applied to extract features' parameter codes including formants, pitch, amplitudes, and duration. The collected text corpus consists of MLT sentences and syllables from MLT books. A Taiwanese balanced sentences speech database including MLT sentences is established through training and analyzing system developed by windows programming, and another rare phoneme units are generated to be included in the database. The phonetic set of Taiwanese tonal phonemes is generated from the HTK recognition results. Through this research, the established TTS system may contribute to the education and training of native Taiwanese for the children, elder, and speech disabled.

This study proposes a bed posture detection method using Bayesian classification for the elderly and bedridden. Only 16 long-narrow FSR (Force Sensing Resistor) sensors, rather than pressure distribution image from a set of sensor array are used for classification. Kurtosis and skewness are estimated as feature vector to represent the shape of pressure contour using the pressure values received from sensors. Gaussian distribution is adopted for statistical modeling and classification for bed postures including supine, left/right lying. Experimental results reveal that proposed method exhibits encouraging potential in bed posture detection.

The paper proposes a maximum a posteriori (MAP) based approach to segment and identify jointly an utterance with mixed languages. A statistical framework for language boundary detection and language identification is proposed. First, the MAP estimation is used to determine the boundary number and positions. Further, an LSA-based GMM and a VQ-based bigram language model are proposed to characterize a language and used for language identification. Finally, a likelihood ratio test approach is used to determine the optimal number of language boundaries. Experimental results show that the proposed approach exhibits encouraging potential in mixed-language segmentation and identification.

This paper proposes a perceptual modeling approach with a two-stage recognition to deal with the issues of recognition degradation in noisy environment. The auditory masking effect is used for speech enhancement and acoustic modeling in order to overcome the model inconsistencies between training speech and noisy input. In the two-stage recognition, the maximum a posteriori (MAP) based adaptation algorithm is used to incrementally adapt the noise model. In order to evaluate our proposed approach, a Mandarin keyword spotting system was constructed. The experimental results show our proposed method achieves a better recognition rate compared to the audible noise suppression (ANS) and parallel model combination (PMC) methods for both in 70km/hr (10.3dB) and 90km/hr (6.4dB) car environments.