Noise Feedback Coding Revisited： Refurbished Legacy Codecs and New Coding Models

Noise feedback coding(NFC) has attracted renewed interest with the recent standardization of backward-compatible enhancements for ITU-T G.711 and G.722.It has also been revisited with the emergence of proprietary speech codecs,such as BV16,BV32,and SILK,that have structures different from CELP coding.In this article,we review NFC and describe a novel coding technique that optimally shapes coding noise in embedded pulse-code modulation(PCM) and embedded adaptive differential PCM(ADPCM).We describe how this new technique was incorporated into the recent ITU-T G.711.1,G.711 App.III,and G.722 Annex B(G.722B) speech-coding standards.     

Mild and severe udder cleft dermatitis—Prevalence and risk factors in Swedish dairy herds

Udder cleft dermatitis (UCD) is an inflammatory skin condition affecting the anterior parts of the udder of dairy cows. The lesions may present as mild or severe skin lesions and have been associated with mastitis and digital dermatitis. The full etiology and pathogenesis are not understood and no large-scale studies have investigated prevalence and risk factors. Therefore, the main objectives of the study were to investigate the prevalence of mild and severe UCD in Swedish dairy herds and to identify risk factors associated with such lesions. We also wanted to investigate risk factors for all cases of UCD and to determine whether UCD increases the risk for mastitis and culling. A random sample of 100 freestall dairy herds were included in the study, and each herd was visited once. Cows were registered as having no, mild, or severe UCD. Additional cow and herd data were obtained via observations, interviews, and the Swedish Official Milk Recording Scheme. The data were analyzed using logistic regression models to identify risk factors for mild and severe UCD. In total, data from 3,479 cows in 99 herds were analyzed. The prevalence of mild and severe UCD was 19 and 9%, respectively. Lesions were found in 98 of 99 herds but the within-herd prevalence of mild (0–43%) and severe (0–33%) UCD varied notably between herds. Breed (Swedish Red compared with Swedish Holstein), certain udder conformation traits, and higher parity were risk factors associated with increased risk of UCD. In addition, cows with hock lesions and cows in herds with high incidence of culling due to hoof and leg diseases had a higher risk for mild UCD. More days in milk and high milk yield were cow-related risk factors associated with severe UCD. Three housing-related factors (shorter cubicles, mattress as cubicle base, and cubicles installed before 2001 compared with 2001–2005), a high incidence of veterinary-treated clinical mastitis and culling due to udder diseases, and a low incidence of culling of first-parity cows in early lactation were herd-related risk factors associated with increased risk for severe UCD. In addition, cows in herds with a high proportion of heifers older than 17 mo that were not inseminated were associated with lower risk of all UCD. Finally, UCD was not associated with the outcomes milk somatic cell count, veterinary-treated clinical mastitis, or culling in the multivariable analyses. The etiology of UCD is most likely multifactorial, involving udder conformation traits and other cow-related risk factors as well as herd-related risk factors. The high prevalence of severe UCD lesions in Swedish dairy cows emphasizes the need for preventive measures and efficient treatments.     

Phonetic segmentation of speech signal using local singularity analysis

This paper presents the application of a radically novel approach, called the Microcanonical Multiscale Formalism (MMF) to speech analysis. MMF is based on precise estimation of local scaling parameters that describe the inter-scale correlations at each point in the signal domain and provides efficient means for studying local non-linear dynamics of complex signals. In this paper we introduce an efficient way for estimation of these parameters and then, we show that they convey relevant information about local dynamics of the speech signal that can be used for the task of phonetic segmentation. We thus develop a two-stage segmentation algorithm: for the first step, we introduce a new dynamic programming technique to efficiently generate an initial list of phoneme-boundary candidates and in the second step, we use hypothesis testing to refine the initial list of candidates. We present extensive experiments on the full TIMIT database. The results show that our algorithm is significantly more accurate than state-of-the-art ones.     

Noise variance estimation based on dual-channel phase difference for speech enhancement

In this paper, we propose a method for estimating a signal-to-noise ratio (SNR) in order to improve the performance of a dual-microphone speech enhancement algorithm. The proposed method is able to reliably estimate both a priori and a posteriori SNRs by exploring a direction-of-arrival (DOA)-based local SNR that is defined by using spatial cues obtained from dual-microphone signals. The estimated a priori and a posteriori SNRs are then incorporated into a Wiener filter. Consequently, it is shown from an objective perceptual evaluation of speech quality (PESQ) comparison and a subjective listening test that a speech enhancement algorithm employing the proposed SNR estimate outperforms those using conventional single- or dual-microphone speech enhancement algorithms such as the Wiener filter, beamformer, or phase error-based filter under different noise conditions ranging from 0 to 20 dB.     

一种利用离散余弦变换频谱图和小波变换图像编码的语音编码算法

语音、图像编码技术是现代数字通信的核心技术,通常是分开进行的,提出了一种利用二维DCT频谱图和小波变换图像编码的语音编码算法.深入分析了算法实现的关键技术——语音信号的二维DCT频谱图,仿真实现了该编码算法,对一段语音的处理结果表明,小波分解系数置零率为83.1941％,压缩后保存能量92.0867％,合成语音达到“优...     

How modality specific is the iambic–trochaic law? Evidence from vision.

The iambic–trochaic law has been proposed to account for the grouping of auditory stimuli: Sequences of sounds that differ only in duration are grouped as iambs (i.e., the most prominent element marks the end of a sequence of sounds), and sequences that differ only in pitch or intensity are grouped as trochees (i.e., the most prominent element marks the beginning of a sequence). In 3 experiments, comprising a familiarization and a test phase, we investigated whether a similar grouping principle is also present in the visual modality. During familiarization, sequences of visual stimuli were repeatedly presented to participants, who were asked to memorize their order of presentation. In the test phase, participants were better at remembering fragments of the familiarization sequences that were consistent with the iambic–trochaic law. Thus, they were better at remembering fragments that had the element with longer duration in final position (iambs) and fragments that had the element with either higher temporal frequency or higher intensity in initial position (trochees), as compared with fragments that were inconsistent with the iambic–trochaic law or that never occurred during familiarization. (PsycINFO Database Record (c) 2011 APA, all rights reserved)     

基于MF-DFA的语音情感识别

针对语音情感线性参数在刻画不同情感类型特征上的不足,将多重分形理论引人语音情感识别中．通过分析不同语音情感状态下的多重分形特征,提取多重分形谱参数和广义hurst指数作为新的语音情感特征参数,并结合传统语音声学特征,采用支持向量机SVM对其进行语音情感识别．试验结果表明,该方法可使系统的准确率和稳定性得到有效提高．非线性参数的引人为语音情感识别提供了一个新的思路．     

Perceptual grouping affects pitch judgments across time and frequency.

Pitch, the perceptual correlate of fundamental frequency (F0), plays an important role in speech, music, and animal vocalizations. Changes in F0 over time help define musical melodies and speech prosody, while comparisons of simultaneous F0 are important for musical harmony, and for segregating competing sound sources. This study compared listeners' ability to detect differences in F0 between pairs of sequential or simultaneous tones that were filtered into separate, nonoverlapping spectral regions. The timbre differences induced by filtering led to poor F0 discrimination in the sequential, but not the simultaneous, conditions. Temporal overlap of the two tones was not sufficient to produce good performance; instead performance appeared to depend on the two tones being integrated into the same perceptual object. The results confirm the difficulty of comparing the pitches of sequential sounds with different timbres and suggest that, for simultaneous sounds, pitch differences may be detected through a decrease in perceptual fusion rather than an explicit coding and comparison of the underlying F0s. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

基于自相关的语音自适应增强器的设计与实现

为了提高语音信号的清晰度,利用基于自相关的自适应滤波器对语音信号进行增强.在理论上研究了该滤波器的频率响应和语音增强原理.同时设计并实现了使用该算法的自适应语音增强器,介绍了该增强器的硬件结构.经过试验研究,证实该增强器能对白噪声环境中的语音信号进行有效增强,实际使用效果明显.     

基于连续识别的嵌入式孤立词识别系统

基于线性网络的孤立词识别系统识别时间与词表规模成正比,识别性能严重受限于词表的规模.根据汉语孤立词特点,提出了一种基于连续识别的大词表孤立词识别系统.该系统围绕嵌入式识别所关注的速度和内存消耗性能,采用先实现应用多级搜索、定点化策略的连续识别,再对连续识别的识别结果进行音字转换处理的方法,将连续语音识别应用于大词表孤立...