Noriyo Hara

Fundamental frequency pattern generator, method and program

According to this fundamental frequency generating method, a fundamental frequency pattern is set from a data base of a fundamental frequency pattern of each accent phrase standardized by the phoneme time length or the time length of the vowel and the vowel corresponding portion, and when the corresponding fundamental frequency pattern is not stored in the data base, the fundamental frequency pattern is generated by interpolating the interval between points serving as the references of the fundamental frequency pattern. With this method, a fundamental frequency pattern having higher naturalness than with conventional methods can be generated.

A multilingual text-to-speech system

An MITalk-based real-time system for unrestricted English, Japanese, and Chinese text-to-speech synthesis has been developed with high observed intelligibility (94% for word recognition in English and 70.8% for 119 Japanese monosyllables). The approach in making the system multilingual has been simple yet effective: in developing the Japanese and Chinese systems, the authors have used the hardware and the program structure and tools of the system originally developed for English, but have otherwise kept the development of the different languages separate. In addition to substantial improvements and modifications of the original MITalk, two major innovations in the system are (1) the method for controlling the glottal pulse and (2) the Phoenix development tools, which combine a powerful and flexible parameter editor with programs for acoustic analysis.<<ETX>>

How can we make IT devices easy for older adults? effects of repetitive basic operation training and help-guidance on learning of electronic program guide system

Older adults have difficulties in using unfamiliar IT appliances because they look poor to learn operations in trial-and-error fashion. To find a different way for older adults to learn operations effectively, an experiment like usability testing with Electronic Program Guide (EPG) system on a HDD/DVD recorder was executed. Unfocused general information was not helpful for learning new operations. Repeated practice of basic operations with timely help-guidance messages facilitated the acquisition of targeted elementary actions and made them easy to learn new elementary actions as well as new complicated operations. It was suggested that older adults could learn better from successful operations than from trial-and-errors, which may induce harmful error spiral. It is necessary to investigate how to avoid error spiral in self-learning environment in home use.

Top-down controlled and optimal syllable-template chaining algorithm for large vocabulary speech recognition

The vowel-consonant-vowel (VCV) template-matching is useful for large vocabulary speech recognition and continuous speech recognition from the point of view of the allophonic variation in phonemes due to co-articulation, the segmentation ambiguity, and the size of vocabulary. In this paper a new matching algorithm for a large vocabulary speech recognition is proposed, and experimental result and some discussion are given. This algorithm has the advantage of moderating an intensive and time-consuming task for template-training by reducing the number of templates to about 40% of the all VCV combinations. Further, by top-down controlled matching, this algorithm has the advantage of reducing computational requirement, which is required to avoid the effect of co-articulation and the segmentation ambiguity.

Effects of a computer‐based voice modification device for a speaker with recurrent nerve paralysis

Patients with unilateral recurrent laryngeal nerve (RNL) paralysis may exhibit severe dysphonia depending on the resting position of the affected vocal fold. Major vocal symptoms of these patients are breathy voice, decreased vocal intensity, and physical fatigue due to great effort to approximate the vocal folds and to increase subglottal pressure. In some cases, phonosurgery is not applicable due to other medical problems, while a good prognosis is not expected by voice therapy. Technological assistance may be needed for such a patient. In this study, a real‐time formant analysis‐synthesis method was used to enhance speech intelligibility of a severely dysphonic patient with unilateral RNL paralysis. The voice source was modified by using inverse‐filtered signals extracted from a normal speaker. A special hardware unit was designed to perform an analysis‐synthesis process with temporal delay of 40 ms for signal processing. Vowels and sentences produced by the subject both with and without the device wer...

Text‐to‐speech system for English and Japanese

A real‐time text‐to‐speech system for English and Japanese has been developed. This system consists of a language processing module, a phonetic acoustic processing module, and a synthesis module. Full general English and Japanese sentences can be converted to speech. The Japanese software and English software are independent except for the synthesis module. The features of this system are as follows. (1) The synthesis module is a phoneme‐based cascade‐parallel formant synthesizer with high observed intelligibility (73.5% for the 119 Japanese monosyllables). (2) This system has a 3000‐morphene English dictionary and 40 000‐word Japanese dictionary with a high‐speed search algorithm. (3) A large speech database was collected for the development of Japanese prosody rules. (4) For the precise control of pitch contour, the Fujisaki model was adopted. (5) One of the two systems developed can stand alone; the other requires a personal computer with a high‐speed DSP board. (6) In the development of this system, s...