For the conventional block floating-point quantizer (BFPQ), usually a large block size causes performance degradation and thus small block sizes are preferred, especially when non-uniformly distributed signals are processed. A tunable BFPQ with fractional exponent is proposed in this paper to deal with the problem. We first examine the root cause of degradation through analytic equations and then propose to tune the thresholds for deriving the exponent and fractional exponent of the block so as to strike a good balance between the quantization error and saturation error. An optimal tuning value depending on the block size and mantissa word-length can be obtained. Thus, the tunable BFPQ can achieve better output signal-to-quantization-noise ratio (SQNR) in a wide dynamic range. The analytic equation for the output SQNR of the proposed BFPQ is derived to verify the simulated results. Only one extra multiplication is required for each block to implement the tunable BFPQ. Finally, we show the obvious SQNR improvements compared to the conventional scheme for various settings of block sizes and mantissa word-lengths.

Four measures were used to observe a hearing-impaired child's volubility, canonical babbling, consonantal development, and syllable complexity. Child A is a seriously hearing-impaired child identified early and provided with a cochlear implant at the age of 14 months. Child B with normal hearing was recruited for the same age range comparison. The recording place was in their homes and the natural interaction between caretakers and children was recorded. The data were collected at the age of 9-30 months. First fifty identifiable utterances in each session were selected for analysis. The indicated that 1) Two children had no significant difference in volubility at pre-linguistic stage. After 21 months, Child A produced more than Child B. 2) The onset of canonical babbling was at as early as 9 months of age in Child B, while Child A had the onset at the age of 18 months. From 27 months, Child A had similar amount of canonical babbling as Child B. 3) Child A had smaller consonantal variability than Child B. 4) Child B had a smooth and gradual growth from using simple syllable structures to more complex forms, whereas Child A showed unstable and sudden shift. These results indicated that Child A had approximately six months phonetic delay. Child B experienced rapid phonetic growth at 18 months, while Child A didn't show obvious growth until 27 months. Child A seemed to make up the delay and narrowed the gap with Child B after the cochlear implantation for one year.

This study aims to examine the influence of lexical tone upon voice onset time (VOT) in Mandarin and Hakka. Examination of VOT values for Mandarin and Hakka word-initial stops in different lexical tones revealed a significant influence. In Mandarin, stops' VOTs, ordering from the longest to the shortest, are in Tone 2, 3, 1, and 4, which conforms to Liu et al.'s (2008) results. However, it was found that the sequence results from the existence of non-words. Participants tended to pronounce non-words at a lower speed in order to produce them correctly. Therefore, data without non-words were further examined, in which no clear sequence had been found. For Hakka, Post-hoc tests show that aspirated stops in Tones 4 and 8 have significantly shorter VOT values than they have in other tones. The result suggests that future studies should consider the effect of lexical tone when studying in tonal languages.

Voice onset time (VOT) has been considered as one of the crucial methods for examining the timing of voicing in stop consonants in many languages. This study was designed to examine VOT production for phonetically voiceless stops of Mandarin and English by native Chinese speakers. Thirty-six Taiwanese Chinese speakers participated in this study. The results indicates that: 1) Based on the three universal categories proposed by Lisker and Abramson (1964), for phonetically voiceless stops, Mandarin and English occupies the same place along the VOT continuum; 2) The mean VOT value for the apical stop /t/ is slightly lower than the mean value for the labial stop /p/. This does not conform to the general agreement, that is, the further back the place of articulation the longer VOT.; 3) The difference between the mean VOT values of English /p/ and /t/ produced by Chinese speakers is subtle, whereas the difference reaches significance for native English speakers. Therefore, it is suggested that first language (i.e. Chinese) could be a crucial factor affecting L2 (i.e. English) productions. Moreover, variations, both within and between speakers, could be further examined in future studies.

This paper considers the problem of estimating the autoregressive moving average (ARMA) power spectral density when measurements are corrupted by noises and with missing data. The missing data model is based on an unknown probabilistic structure. In this situation, the spectral estimation becomes a highly nonlinear optimization problem with many local minima. In this paper, we use the global search method of genetic algorithm (GA) to achieve a global optimal solution of this spectral estimation problem. From the simulation results, we have found that the performance is improved significantly if the probability of data missing is considered in the spectral estimation problem.

In the presence of narrowband interference (NBI), the performance of the direct-sequence code-division multiple access communication system can be improved by using various interference cancellation schemes. In this paper, a scheme of parallel interference cancellers (PICs), which includes a set of NBI cancellation filters and a set of multiple access interference (MAI) cancellation filters, is developed to improve the signal-to-interference ratio (SIR) and handle the MAI effect. In order to avoid performance deterioration due to unreliable initial detection, a robust coefficient /spl gamma/ is introduced in our proposed design criterion. A recursive least-square algorithm is employed to update the tap coefficients of the PICs. In other words, no training or retraining period is needed in our proposed PICs. Finally, the SIR performance of the proposed detector is analyzed and compared to that of the conventional linear detector.