语音的

On basis of analyzing discriminative characters of speech and music signals, a real-time speech/music classification arithmetic is presented, which classifys audio signal into speech and music in real-time with tonality distribution property and spectral tilt Additionally, the classification results are smoothed for rationality.

分析了语音和音乐的区别性特征，提出了一种利用音调分布特性和频谱倾料度对语音和音乐进行实时分类的算法，并对分类结果进行简单平滑，使分类更为合理，从而提高音质。

Phrase recognition in fixed phrase table, with many applications, is an important research area of speech recognition.

摘要］ 具有确定词表的词组语音识别是语音识别研究的一个重要方面，应用相当广泛。

Keyword spotting is an important area in speech recognition. Its objective is to identify and verify a few specified key-words in continuous speech.

关键词识别是语音识别研究中的一个重要领域，其目的是从连续语音中检测并确认给定的若干个特定关键词。

A feasible method is introduced to code and decode(converting between linear code and A-law pulse-code-modulation code)in order to reduce the quantity of speech data,and to superpose speech signal.

目前，51单片机应用广泛、性价比高，它的位处理功能在语音信号处理中也有很大优势，很多产品研发人员希望以单片机为核心研制经济实惠的语音系统。

English speakers and the college English majors in the top three funny jokes and in the top three unfunny jokes in linguistic jokes are that Syntactic Joke 8 was favored by both groups; Phonological Joke 3 was not favored by both groups; the two Phonological Jokes were are not appreciated by the NSs; Lexical joke 15 is not favored by the CEMs, but is appreciated most by the NSs; and Morphological Joke 5 is not appreciated by the CEMs, but is favored by the NSs.

在语言笑话欣赏中，2组受试者分别在其前三名好笑的和前三名不好笑的笑话排行比较结果显示(1) 2组受试者都觉得句法笑话8好笑；(2) 2组受试者都觉得语音笑话3不好笑；(3)英语外籍人士觉得2则语音笑话都不好笑；(4)台湾学生觉得字汇笑话15不好笑，但是英语外籍人士觉得最好笑；(5)台湾学生觉得词素笑话5不好笑，但是英语外籍人士觉得好笑。

A novel approach for isolated digit speech recognition based on simplified pulse coupled neural network is presented in terms of the visible characteristics of speech.

根据语音信号的"可视"特点，提出了一种基于简化脉冲耦合神经网络实现孤立词语音识别的方法。

At the present time, EVRC is the best vocoder in the CDMA system when take into account both the voice quality and the encode rate.

在目前的CDMA系统中，综合语音质量和编码速率，EVRC是最佳的语音编码器。

Segment vocoder is an efficient very low bit rate speech coding algorithm.Speech segment method,one of the key technology of segment vocoder,is studied and a new algorithm is presented.

分段声码器是一种高效的低速语音编码方法，针对分段声码器实现的关键技术语音分段方法进行了研究，提出了一种分段新算法。

Primary phonological iconicity is sub-divided into auditory iconicity, articulary iconicity and associative iconicity.

而次要语音象似主要是指人在交流中的语音语调符合语境和说话者的心情态度等。

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。

She gently rebuff ed him, but agreed that they could be friends

她婉言拒绝了，但同意作为朋友相处。

If in the penal farm, you were sure to be criticized.

要是在劳改农场，你等着挨绳子吧！