语音

P＜0.05.conclusion centralized,systematic and individual speech training may help the postoperative patientswith well velopharyngeal closure feckly improve their speech articulation to normal.

集中系统并辅以个性化语音训练，可以使腭裂术后腭咽闭合功能良好的患者，语音清晰度提高到与正常人基本一致。

Results In resonance, 52 patients (80.0%) were hypernasality, 13(20.0%) were hyponasality, and in articulation, 47 patients (72.3%) were omission, 19 (29.2%) were backing, 16 (24.6%) were fronting, and 32 (49.2%) were glottal.

结果 语音评估结果显示，共鸣方式为高鼻音52例（占80.0%），低鼻音13例（占20.0%）；从音韵结构历程和病人构音位置等方面分析，声母省略47例（占72.3%），声母舌后置19例（占29.2%），声母舌前置16例（占24.6%），而喉头音和代偿性语音32例（占49.2%）。

Objective and subjective tests show that the quality of speech synthesized by the improved method is superior to the one synthesized by the IMBE algorithm with adaptive forward-backward quantization, and as good as the one synthesized by the 2.4kbps IMBE algorithm.

主客观测试表明，改进算法的合成语音质量好于自适应前后向量化IMBE算法，且合成语音质量与 2.4kb/s IMBE算法相当。

AMR, which is belonging to CELP, established by 3GPP, is the speech compressing coding algorithm applied in the 3G mobile communication system-WCDMA system, resolving the rate assignment problem between the source coding and channel coding much more intellectively, making the allocation and utilization of wireless resource much more active and efficiently.

AMR（自适应多速率语音编码，Adaptive Multi Rate）是由3GPP制定的应用于第三代移动通信W-CDMA系统中的语音压缩编码，更加智能地解决了信源和信道编码的速率分配问题，使得无限资源的配置和利用更加灵活和高效。

In my memory,my deskmate in senior middle school was a girl in a long dress,standing in early June breezes,singing gently her own sentimental verses about her favorite trellis vines flower,her innnermost feelings...with her hairpins glittering and kissingly swinging,just like a young lady from a

日语入门的基础日语语音是日语入门的基础，这个想必大家都非常清楚，在这里我希望大家在学习的过程中，不要只求速度，这样很容易返工的，不如在开始的时候多下点儿工夫多记语音。

Correct unit segmentation are, though laborsome, very crucial to the performance of a concatenation based TTS system.

语音单元边界的准确切分对基于波形拼接的语音合成系统至关重要。

Finally, based on the hybrid method, a speaker independent Mandrin continuous speech recognition system that convert speech into Mandrin pinyin is developed.

此外，基于HMM/MLP混合方法建立了一个把语音转换成拼音的非特定人汉语连续语音识别系统。

A PSOLA based waveform coding synthesis technology in the Chinese TTS system is proposed and implemented to approve the speech's naturality and the auditory effect

在此基础上，提出用基于PSOLA的波形编辑语音合成技术应用于汉语TTS系统中，以提高输出语音的自然度和听觉效果。

A new algorithm for digital audio watermarking is proposed in this paper. This algorithm bases on the wavelet transform. By using this algorithm, the watermarked speech signal sounds the same as the original one. The tiny difference is neglectable by human ear.

主要研究了在数字语音信号中加入数字水印的方法，提出了一种基于小波变换的数字水印隐藏与检测算法，用这种算法隐藏水印具有很强的隐蔽性，对原始语音的影响基本上察觉不出来。

Voice signals in the frequency domain processing, voice is a time-varying, nonstationary random process.

语音信号的频域处理，语音虽然是一个时变、非平稳的随机过程。

However, as the name（read－only memory）implies, CD disks cannot be written onorchanged in any way.

然而，正如其名字所指出的那样，CD盘不能写，也不能用任何方式改变其内容。

Galvanizes steel pallet is mainly export which suits standard packing of European Union, the North America. galvanizes steel pallet is suitable to heavy rack. Pallet surface can design plate type, corrugated and the gap form, satisfies the different requirements.

镀锌钢托盘多用于出口，替代木托盘，免薰蒸，符合欧盟、北美各国对出口货物包装材料的法令要求；喷涂钢托盘适用于重载上货架之用，托盘表面根据需要制作成平板状、波纹状及间隔形式，满足不同的使用要求。