共振峰

LPC方法提取语音信号共振峰的分析

Analysis of Speech Signal Formant Abstracted by LPC

早期的TTS系统多采用参数合成方法，如共振峰合成和LPC参数合成。

The early TTS system often employs synthesis methods based on parameters , such as formant synthesis or LPC synthesis .

普通话双音节中第二音节辅音对第一音节韵母／u／共振峰轨迹的影响

The Variety of Formants Tracks of Final / u / of the First Syllable in Disyllables of Mandarin

最小均方误差（LMS）自适应算法用来进行语音信号的共振峰跟踪。

The LMS adaptive algorithm is used for tracking formants of speech signals .

孤子附近出现了一个新的局域振动模G6；由于表面局域态的存在，声波的透过率出现共振峰。

A new localized mode have been found ; The particular resonance peak in transmissivity function is found due to the surface localized mode .

溶剂共振峰（水峰）的抑制是核磁共振（NMR）波谱学，特别是生物NMR研究中的关键技术之一。

Solvent resonance suppression is one of the important techniques in NMR spectroscopy , especially in biological NMR studies .

除a外，正常儿童的各单元音均有完全腭咽闭合，各单元音第三共振峰频率值均高于腭裂组（P＜0.05或P＜0.01）；

The value of F 3 in Chinese vowels of normal children was significantly higher than those of patients with cleft palate except ( P < 0.05 or P < 0.01 );

随着R的增大，共振峰的位置在ω轴上左移，幅值下降，个数增多；

With the increase of , the positions of resonance peaks move towards left , the number of peaks increased , and the amplitude of peak decreased .

基于LPC汉语信号共振峰分析

Resonance Peak Analysis of Chinese Speech Signal Based on LPC

基于Hilbert-Huang变换的语音信号共振峰频率估计

Speech formant frequency estimation based on Hilbert-Huang transform

苯乙烯在TS-1分子筛上的氧化反应存在两条竞争途径：自由基和环氧化。在归属核磁共振峰的基础上，我们提出自由基反应机理。

There are two competing reaction pathways for styrene oxidation over TS-1 zeolite : radical and epoxidation .

病灶NAA，Cr和Cho的相对值以病侧和对侧各代谢物共振峰谱线下积分面积的比值来表示。

The signals from NAA , Cr , and Cho in the lesions were compared with those on the contralateral side as peak area ratios .

通过改变复合颗粒外层SiO2厚度，可以调节其等离激元共振峰的位置；

The plasmon resonance peaks of the resulting SiO2 @ Au @ SiO2 could be controlled by changing the thickness of the outer silica shells .

本文基于线性预测技术（LPC），对语音信号的共振峰参数估值，实现语音合成。

This article provides a method of realizing speech synthesis based on LPC , by using resonance peak parameter estimation .

峰值电流模同步降压型DC-DC变换器设计不同频率共振峰在话者识别中的鉴别价值

The Design of the Synchronous Rectification Buck DC-DC Converter Based on Peak Current-Mode Research on the discriminating value of different frequency formants in speaker recognition

一种基于共振峰恢复和Mellin变换的非特定人语音特征提取方法

A Novel Speaker Normalization Method Based on Formant Recovery and Mellin Transform

通过分析共振峰随空间和能量的变化情况，我们证明该中心处的共振峰与Dirac点是共存的。

By probing its spatial and energy dependent STS , we demonstrate the coexistence of Dirac point and resonance peak at the impurity center .

应用固体1HMASNMR实验技术，根据核磁共振峰线宽的不同，表征了水在不同吸附质中的微观状态。

By using solid state 1H NMR technology , water states on different adsorbent materials in micromolecular level are characterized by the difference of their 1H MAS NMR line width .

在其UV-Vis光谱中只观察到一个位于单金属银和金之间的等离子体共振峰，表明Au-Ag合金纳米粒子已经形成。

One plasmon band between that of monometallic Ag and Au observed in their UV-Vis spectra indicated the formation of Au-Ag alloy nanoparticles .

通过对LPC（线性预测编码）方法提取语音信号共振峰进行的研究表明，采用相－频特性与对数幅－频特性同样能提取语音信号共振峰。

The study on speech signal formant abstracted by LPC shows that it can also be abstracted by adopting phase-frequency characteristics and logarithmic amplitude-frequency characteristics .

除了常用的最低的三个共振峰外，F2／F1、F3／F2可能会给单元音的细节辨识提供两个新的特征参量。

For the discrimination of pure vowels , F2 / F1 and F3 / F2 are possibly the two new parameters besides the three lowest formants .

运用米（Mie）理论和德拜（Doyle）方法，根据测得的等离子共振峰的光宽度，估算出Ag胶态粒子的半径约为18（？）。

The Ag colloid of 18 A in radius was estimated from measurements of the half-width of the plasma resonance by using the Mie theory and Doyle method .

研究了语音合成技术的原理以及发声合成、LPC合成和共振峰合成等技术。

The main researches are follows : 1 . This thesis researches on speech synthesis theory , articulatory synthesis technology , LPC synthesis and formant synthesis technology .

与光滑的直通道相比，狭缝中部的凹口使得奇数的F-P模的共振峰红移，而是偶数的F-P模的共振峰蓝移。

Compared with smooth straight channel , the cut in the middle of the slit makes odd F-P resonance mode red shift and even F-P resonance mode blue shift .

随着温度的升高，Kondo共振峰和电导谷慢慢消去，而总电导随之逐渐增加。

As the temperature increases , the Kondo resonance peak and the conduc - tance dip fade out , and the total conductance increases .

铁磁共振峰峰线宽的变化与共振场变化有明显的不同，当外加磁场随着方位角θH变化时峰峰线宽表现出单峰性，并且在共振场最大值的角度位置却表现出极小值。

However , the variation of FMR linewidths was obvious different from that of resonance fields for all of the films . FMR linewidths revealed the unimodality with the variation of magnetic fields versus the out-plane angles and occurred as minimum at the location of maximal resonance field .

方法对无牙颌患者戴全口义齿前、后与正常对照组进行语音a/i/u的检测并对其共振峰参数频率值（F）、能量值（P）的USSA语音频谱的分析。

Methods Testing patients ' pronunciation with of without complete denture and contrast with normal speech sounds . Using USSA to analyes its frequency and parameter number of wave crest ( F ) and energy number ( P ) .

LPC在非基音同步情况下是一种有偏估计算法，用于共振峰参数估计时误差较大。辅助变量法消除估计偏倚的验证

LPC gives biased estimation when analysis is pitch-asynchronous , so that the errors of formant parameter estimation of LPC are large . Modification of Biased Estimation for Static Single Bearings-only Observer by Instrumental Variables

结果：元音/a/、/i/、/u/共振峰频率（F1，F2，F3）及第一共振峰带宽（B1）各组数据无统计学差异（P>0.05）。

Results : There was no difference ( P > 0.05 ) in vowels formant ( F1 , F2 , F3 ) and the first formant band width ( B1 ) of each group data .

提出直通道一侧有垂直凹口的金属狭缝阵列并利用FDTD方法研究了其透射特性，结果表明：在我们考虑的波长范围内共存在四个高阶F-P模共振峰。

Metallic slit arrays of straight channels with perpendicular cuts are proposed and their transmission properties are investigated numerically using FDTD method . Results indicate that there are four high order F-P resonance modes in our considered region .