Chinese translation for "非特定人"
|
- speaker independent
Related Translations:
非特化定律: law of non-specialization 非特定光源: non-motivational lighting 非特异性抑制: nonspecific inhibition
- Example Sentences:
| 1. | Embedded speaker - independent continuous speech recognition system based on mpc
5200的嵌入式非特定人连续语音识别系统 | | 2. | A robust feature - extraction method based on wavelet transform for text - independent speaker identification
变换的非特定人语音特征提取方法 | | 3. | Speaker - independent continuous speech recognition based on the biomimetic pattern recognition
基于仿生模式识别的非特定人连续语音识别的研究 | | 4. | A baseline system of keyword spotting based on continue hidden markov model ( chmm ) is constructed
研究的主要内容包括: 1 .基于连续隐马尔可夫模型( chmm )框架的非特定人关键词识别基线系统的构建。 | | 5. | Aimed at the problem of single word speech recognition , the paper construct a speech recognition artithmetic based on ann , which applies ann in speech recognition and gains a satisfactory effect
摘要针对非特定人的孤立词识别问题，研究构造了基于神经网络建模的语音识别算法。 | | 6. | An improved clustering method is provided which is based on radial basis function neural network , and applied rbf neural network into the speech sounds recognition domain , and a speaker - independent isolated words speech recognition system is established
摘要提出一种基于径向基函数神经网络的改进聚类方法，并将此改进的神经网络应用于语音识别领域，建立一个非特定人的孤立词语音识别系统。 | | 7. | Following is the main work of this thesis : 1 . after researching on the basic theory of speech recognition , this thesis realizes talker - independent recognizer and speaker identification system with dtw ( dynamic time warping ) and vq ( vector quantization ) algorithm respectively
本文的主要研究工作如下: 1 .在研究语音识别的一些基本理论基础上,分别采用dtw和vq两种识别算法实现了非特定人孤立词识别系统和说话人辨认系统。 | | 8. | Although there still exist many problems which need to be overcome such as the technique of large - vocabulary , speaker - independent , continuous speech recognition , the technique ( lpc and dtw ) for small - vocabulary , speaker - dependent , isolated - word speech recognition has come to the age of maturity
虽然大字表非特定人连续语音识别技术仍有许多难点尚未解决,但特定人小字表孤立词语音识别技术( lpc和dtw算法)已相对成熟。 | | 9. | Improvement has been proposed to provide speaker identification system a better performance . the experiments show that the recognition rate is upper to 96 percent in quiet environment , the recognition rate declines sharply in low snr environment , and with certain low input snr level , nearly unrecognizable . 2
同时,对基于dtw算法的非特定人孤立词识别系统进行了多种噪声环境下的仿真实验,结果表明:在安静环境下,系统识别率可到达96 %以上,而在低信噪比环境下,语音识别系统性能急剧下降,在信噪比极低时,几乎无法识别。 | | 10. | The first part of the paper is designing the testing project for grounding resistance and insulation resistance in a new way . using 16bits ad converter with programmable control amplifier replaced the way which used changing resistance to change measure range . lt is not only improved testing precision and develop the system expediently , but also reduced the area of the circuit boardwith the new way . in order to make the electric implement safety testing system have upstanding expansibility , the software and hardware of the system adopted the modularization design . adopted mcu atmegal28 as a master mcu which control mmi , realtime clock and communication with slaver mcu . atemga8 as the slaver mcu to realize testing function . so it is easy to add or reduce the testing project . the testing implement system has been developed successfully , and the comments for the system is that it has high precision , high expansibility and easy maintain . but considering the electric implement system should have intelligence and humanity abi lity . so this paper bring forward a scheme of electric equipment safety testing embedded system with speech control . after introduce the basic theory of speech recognition , the paper expatiate the characters of this system . the system is a noise conditon , not special people , small glossary , insulation word system . with these characters design the speech recognition as fellow . utilizing cross zero ratio and short energy to ensure jumping - off point and end point ; adopting mfcc as the character parameters of speech recognition ; the character parameters than be recognized by dtw . in order to ensure the credibility of this project , first realized by matlab in computer
在介绍了语音识别的基本原理后，阐述了本系统的特点：本系统是一个噪声环境下非特定人、小词汇量、孤立词的语音识别系统。根据本系统的这些特点设计了如下语音识别方案：利用过零率和短时能量相结合的方式确定语音端点；采用mel频率倒谱系数( mfcc )作为语音识别的特征参数；得到的特征参数最后通过动态时间规整( dtw )的模式识别方法进行识别。为了确保本系统实现方案的可靠性，首先通过计算机利用matlab软件来仿真，在算法仿真实现后又进一步增加环境的复杂性：加上较大的环境噪声、突发性的噪声等，再通过修改参数、修改参考模板、两级识别等各种提高语音识别精度的方法来提广东工业大学工学硕士学位论文高识别率。 |
- Similar Words：
- "非特定订单" Chinese translation, "非特定反应" Chinese translation, "非特定关系" Chinese translation, "非特定光源" Chinese translation, "非特定迁移" Chinese translation, "非特定条件下贮存" Chinese translation, "非特定物" Chinese translation, "非特定语者语音识别" Chinese translation, "非特化定律" Chinese translation, "非特化法则" Chinese translation
|
|
|
