检索结果-内蒙古大学图书馆

您好，读者！请登录

内蒙古大学图书馆

首页
概况
党建
资源
服务
科研支持
- 论文收录引用证明
- 科技查新
知识产权
档案馆
帮助

咨询与建议

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

您的常用邮箱：*

您的手机号码：*

问题描述：

当前已输入0个字，您还可以输入200个字

全部搜索
期刊论文
图书
学位论文
标准
纸本馆藏
外文资源发现
数据库导航
超星发现

高级检索

时间限定

出版年份：

文献类型

图书期刊文献学位论文多媒体

馆藏选择

电子馆藏纸本馆藏

核心期刊

全部期刊 SCI 收录期刊 SSCI 收录期刊 EI 收录期刊 CSCD 收录期刊 CSSCI 收录期刊

语言

中文英文

文献类型

期刊文献图书学位论文标准纸本馆藏

帮助

文字说明：

T=题名（书名、题名），A=作者（责任者），K=主题词，P=出版物名称，PU=出版社名称，O=机构（作者单位、学位授予单位、专利申请人），L=中图分类号，C=学科分类号，U=全部字段，Y=年（出版发行年、学位年度、标准发布年）

检索规则说明：

AND代表“并且”；OR代表“或者”；NOT代表“不包含”；(注意必须大写,运算符两边需空一格)

检索范例：

范例一：(K=图书馆学 OR K=情报学) AND A=范并思 AND Y=1982-2016
范例二：P=计算机应用与软件 AND (U=C++ OR U=Basic) NOT K=Visual AND Y=2011-2016

分类表

所选分类

>> <<

限定检索结果

文献类型

1,158 篇 会议
12 篇 期刊文献
1 册 图书
1 篇 报纸

馆藏范围

1,172 篇 电子文献
0 种 纸本馆藏

日期分布

学科分类号

701 篇 工学
- 272 篇 信息与通信工程
- 188 篇 仪器科学与技术
- 151 篇 电子科学与技术（可...
- 136 篇 控制科学与工程
- 135 篇 计算机科学与技术...
- 94 篇 软件工程
- 89 篇 网络空间安全
- 46 篇 机械工程
- 26 篇 光学工程
- 13 篇 航空宇航科学与技...
- 6 篇 兵器科学与技术
- 5 篇 测绘科学与技术
- 4 篇 公安技术
- 3 篇 材料科学与工程（可...
- 3 篇 地质资源与地质工...
- 2 篇 电气工程
- 2 篇 矿业工程
505 篇 理学
- 450 篇 系统科学
- 39 篇 数学
- 16 篇 物理学
- 2 篇 地理学
- 2 篇 统计学（可授理学、...
83 篇 军事学
- 82 篇 军队指挥学
- 5 篇 战术学
59 篇 管理学
- 56 篇 管理科学与工程(可...
- 3 篇 工商管理
59 篇 艺术学
- 59 篇 设计学（可授艺术学...
12 篇 文学
- 9 篇 中国语言文学
- 2 篇 外国语言文学
10 篇 经济学
- 10 篇 应用经济学
2 篇 法学
2 篇 农学
2 篇 医学
1 篇 教育学

主题

71 篇 语音识别
35 篇 声学模型
22 篇 神经网络
21 篇 编码
21 篇 信息论
20 篇 译码
18 篇 语音合成
18 篇 算法
17 篇 hmm
13 篇 隐马尔可夫模型
12 篇 端点检测
12 篇 移动通信
11 篇 非特定人
11 篇 鲁棒性
11 篇 信噪比
11 篇 数据库
10 篇 码分多址
10 篇 通讯理论
10 篇 性能分析
10 篇 学术会议

机构

89 篇 清华大学
72 篇 北京邮电大学
38 篇 中国科学院声学研...
35 篇 北京大学
31 篇 电子科技大学
30 篇 北方工业大学
28 篇 哈尔滨工业大学
25 篇 中国科学技术大学
24 篇 北京邮电学院
19 篇 南京航空航天大学
18 篇 邮电大学
17 篇 装备指挥技术学院
16 篇 北京交通大学
15 篇 西安电子科技大学
13 篇 东南大学
12 篇 北京航空航天大学
12 篇 云南大学
12 篇 中国科学院自动化...
10 篇 浙江大学
10 篇 中山大学

作者

31 篇 王作英
31 篇 杨义先
30 篇 徐波
22 篇 王仁华
21 篇 吴文虎
17 篇 迟惠生
17 篇 方棣棠
16 篇 黄泰翼
14 篇 郑方
13 篇 颜永红
13 篇 吴玺宏
11 篇 吕士楠
11 篇 莫福源
11 篇 张树武
11 篇 肖熙
10 篇 韩纪庆
10 篇 俞铁城
10 篇 钮心忻
10 篇 李成荣
10 篇 李昌立

语言

1,114 篇 中文
58 篇 英文

检索条件"任意字段=中国通信学会信息论与通信理论学术会议"

共 1172 条记录，以下是1151-1160 订阅

全选清除本页清除全部题录导出标记到"检索档案"

详细简洁

排序：

COMPARISONS AMONG FOUR STATISTICS BASED METHODS OF PROSODY STRUCTURE PREDICTION

COMPARISONS AMONG FOUR STATISTICS BASED METHODS OF PROSODY S...

引用

第七届全国人机语音通讯学术会议（NCMMSC7）

作者： Qin Shi, Wei Zhang, XiJun Ma, WeiBin Zhu, Ling Jin IBM China Research Lab, Beijing, P.R.China

Prosody structure prediction plays an important role in text-to-speech (ITS) conversion systems, ft is the must and prior step to parametric prosody prediction. Dynamic programming (DP) and decision tree (DT) are widely used for prosody structure prediction [1][2][3] but with well-known limitations. In this paper, two other new methods, combination of dynamic programming with decision tree and combination of decision tree with finite state machine (FSM), are proposed. Then, based on a manually labeled corpus, comprehensive comparisons among the four methods are done. It could be concluded from these experiments that combination of dynamic programming with decision tree method is the best choice for prosody word boundary prediction and combination of decision tree with FSM is the best candidate for prosody phrase boundary prediction.

关键词： COMPARISONS AMONG FOUR STATISTICS BASED METHODS OF PROSODY STRUCTURE PREDICTION

来源：评论

学校读者我要写书评

暂无评论

Comparative Study on Different Strategies of Integrating Tonal Information into Mandarin Name Recognition

Comparative Study on Different Strategies of Integrating Ton...

引用

第七届全国人机语音通讯学术会议（NCMMSC7）

作者： Luo Dongsheng, XieXiang, Kuang Jingming Dept. of Electronic Engineering, Beijing Institute of Technology No .5 Zhongguancun Nandajie, Haidian District, Beijing, 100081, P.R.China

As Chinese is well known to be a tonal language, how to apply tonal information is a special issue in Mandarin ASR. In this paper we investigate three different strategies of integrating tonal information based on a multi-stream model. An evaluation is made for a Chinese name recognition task, in which the vocabulary is composed of 100 tonally confusing pairs. Our experimental results show that a significant improvement in the recognition accuracy can be achieved with any combination strategy, and a maximal reduction of 83.2% in word error rate (WER) is obtained by hypothesis combination. Generally, hypothesis combination seems to be the best solution in both recognition accuracy and flexibility for isolated word speech recognition.

关键词： Comparative Study on Different Strategies of Integrating Tonal Information into Mandarin Name Recognition

来源：评论

学校读者我要写书评

暂无评论

A Continuous Speaker-Independent Putonghua Dictation System

A Continuous Speaker-Independent Putonghua Dictation System

引用

第四届全国人机语音通讯学术会议

作者： R．A．Gopinath M．D．Monkowski M．A．Picheny IBM Research Division T．J.Watson Research Center PO Box 218 Yorktown Heights New York 10598 USA IBM Research Division T．J.Watson Research Center PO Box 218 Yorktown Heights New York 10598 USA IBM Research Division T．J.Watson Research Center PO Box 218 Yorktown Heights New York 10598 USA

We describe new methods for continuous putonghua speech recognition. We have augmented the IBM HMM-based continuous speech recognition system with the following features: First, we treat tones in putonghua as attribu... 详细信息

We describe new methods for continuous putonghua speech recognition. We have augmented the IBM HMM-based continuous speech recognition system <1-3> with the following features: First, we treat tones in putonghua as attributes of certain phonemes, instead of syllables. We call those phonemes with tone tonemes. Second, instantaneous pitch is treated as a variable in the acoustic feature vector, in the same way as cepstra or energy. Third, by designing a set of word-segmentation rules to convert the continuous Chinese text into segmented text, the trigram language model works effectively. By applying those new methods, a speaker-independent, very-large-vocabulary continuous putonghua dictation system can be constructed.

关键词： A Continuous Speaker-Independent Putonghua Dictation System

来源：评论

学校读者我要写书评

暂无评论

Speaker Segmentation Based on Model Scoring

Speaker Segmentation Based on Model Scoring

引用

第八届全国人机语音通讯学术会议

作者： Gaoxiong Yi, Qiying Li, Zhongsong Lin, Xihong Wu, Huisheng Chi Speech and Hearing Research Center National Laboratory on Machine Perception Peking University, 100871

A new method of speaker segmentation based on model scoring is proposed in this paper. This method consists of three phases: (1) pre-segmentation, (2) model scoring, and (3) clustering. In the first phase, we divide the utterances into small fragments, assuming that in each part there is only one speaker. In the model scoring phase, the spectrum features extracted from those small speech fragments are mapped to score features by calculating the likelihood over a series of cohort speaker models. In the third phase, we merge the parts which have the smallest distance between each other to form the utterance for one speaker. After that, higher accuracy can be achieved if resegmentation is performed. Results from the experiments in this paper demonstrate that this method indeed produces satisfying accuracy on speaker segmentation.

关键词： Speaker Segmentation, Model Scoring

来源：评论

学校读者我要写书评

暂无评论

Investigating F0 Reset and Range in Relation to Fluent Speech Prosody Hierarchy

Investigating F0 Reset and Range in Relation to Fluent Speec...

引用

第八届全国人机语音通讯学术会议

作者： Chiu-yu Tseng, Chun-Hsiang Chang and Zhao-yu Su Institute of Linguistics, Academia Sinica Taipei, Taiwan 115

We studied FO reset and FO range change in relation to a hierarchical multiple-phrase prosody framework for fluent speech that accounts for cumulative fluent speech output. Both FO reset FO range modifications were analyzed with respect to levels of boundary breaks within and across phrases. We found that FO reset occurred after higher levels of boundary breaks;FO range may reduce for transitional phrases that occurred at the beginning of a speech paragraph, causing the first FO reset to move forward to the next phrase. We also found gender related differences in speaking style and overall global FO contour patterns. The findings are instrumental to enhance both our prosody framework and application to synthesis output.

关键词： Fluent Investigating F0 Reset and Range in Relation to Fluent Speech Prosody Hierarchy

来源：评论

学校读者我要写书评

暂无评论

Discriminative Training for Speaker Identification with Maximum Model Distance Algorithm

Discriminative Training for Speaker Identification with Maxi...

引用

第七届全国人机语音通讯学术会议（NCMMSC7）

作者： Q. Y. Hong and S. Kwong Dept of Computer Science, City University of Hong Kong

Gaussian Markov model (GMM) has been widely used in the area of speaker recognition. It can be considered as a one-state hidden Markov model (HMM) with multiple mixtures. As for the HMM, maximum-likelihood (ML) estimation is considered a good choice of training approach for the GMM-based speaker identification. However it only considers the likelihood of a single speaker. That is, each speaker model is estimated separately using its labeled training utterances. To compare the likelihood against those similar speakers and maximizes their likelihood differences, another training algorithm named maximum model distance (MMD) is proposed for GMM in this paper. Experimental results from the TI46-Word and TIMIT database show that it is an attractive alternative for the GMM training.

关键词： TIMIT GMM Discriminative Training for Speaker Identification with Maximum Model Distance Algorithm MMD Gaussian Model

来源：评论

学校读者我要写书评

暂无评论

EARSCTS:A Chinese Telephony Conversational Corpus for Speech Processing

EARSCTS:A Chinese Telephony Conversational Corpus for Speech...

引用

第八届全国人机语音通讯学术会议

作者： Pascale Fung Christopher Cieri Human Language Technology Center Department of Electrical and Electronic Engineering University of Science and Technology Hong Kong Linguistic Data Consortium University of Permsylvania U.S.A

This paper describes the development and collection as well as initial analysis of the EARS (Effective, Affordable, Reusable Speech-to-text) Chinese telephony speech corpus (EARSCTS). The corpus contains 1206 ten-minute natural Mandarin conversations between either strangers or friends. The total amount of the corpus is 200 hours. There are 40 topics in all conversations and each conversation focuses on a single topic. All the speech data are recorded over public telephone networks, e.g., landline and cellular channels. All the speech data are annotated manually with standard Chinese characters (GBK) as well as specific mark-ups for spontaneous speech, time alignment information is also provided. This corpus can be used for conversational and spontaneous Mandarin speech recognition and other application-dependent tasks. The EARSCTS corpus is the largest and first of its kind for Mandarin conversational telephony speech, providing sufficient and diversified samples needed for speech training, testing, adaptation and development.

关键词： EARSCTS:A Chinese Telephony Conversational Corpus for Speech Processing

来源：评论

学校读者我要写书评

暂无评论

INSIGHT INTO THE ROLE OF PITCH INFORMATION IN TEXT-INDEPENDENT SPEAKER RECOGNITION

INSIGHT INTO THE ROLE OF PITCH INFORMATION IN TEXT-INDEPENDE...

引用

第八届全国人机语音通讯学术会议

作者： Xuefeng Zhang, Yuan Dong, Xianyu Zhao, Hao Yang, Haila Wang France Telecom Research & Development Center, Beijing, China, 100083 Beijing University of Posts and Telecommunications

This paper provides explanation and insight regarding the role of pitch information in text-independent speaker verification by experiments. The paper first brief a robust fundamental frequency extraction method. Then, a channel robust text-independent speaker verification based on GMM-UBM model which fuse the extracted pitch information and acoustic features is described in detail. Experiments are conducted using the NIST 1999 Speaker Recognition Evaluation corpus. Experimental results are presented show that the pitch features are more robust to handset variability and noise, but are not very robust to other factors like emotions. Therefore, how to effectively utilize the pitch information in speaker recognition is still a challenge for researchers.

关键词： PLP INSIGHT INTO THE ROLE OF PITCH INFORMATION IN TEXT-INDEPENDENT SPEAKER RECOGNITION GMM

来源：评论

学校读者我要写书评

暂无评论

Towards Hierarchical Speech Recognition

Towards Hierarchical Speech Recognition

引用

第七届全国人机语音通讯学术会议（NCMMSC7）

作者： Mingxing Xu, Dali Yang, Wenhu Wu State Key Laboratory of Intelligent Technology and Systems Department of Computer Science and Technology Tsinghua University, Beijing, 100084, China

Hierarchical recognition has been proposed for a long time in the pattern recognition field. Although it is a familiar action when human performs a recognition task, there is not an effective and systematic method to implement it for the speech recognition. This paper presents our recent experimental results on this topic, which uses the principle of sub-space partition to realize a hierarchical recogntion and a tree-based architecture to organize multi-recognizers. Although these preliminary evaluations are carried out with an isolated word recognition system for 10 Chinese digital syllables, we believe that it is easy to extend them for the spontaneous speech recognition. The results show that the proposed algorithm can achieve about 10% error reduction compared with traditional methods. In future works, we will test all Chinese syllables and extend them for the continous speech recogntion.

关键词： Towards Hierarchical Speech Recognition

来源：评论

学校读者我要写书评

暂无评论

Audio Search for Spoken Documents Content-Based Retrieval

Audio Search for Spoken Documents Content-Based Retrieval

引用

第八届全国人机语音通讯学术会议

作者： Romain Le Tellier Signal and Image Processing department GET-ENST Paris France

The amount of digital audio and video documents being shot and stored in large archives is growing faster than ever before. However, lack of effective content searching methods is a major barrier that prevents people from operating audio and video databases pervasively and intelligently. This paper reviews the state-of-the-art in audio information retrieval and presents an audio searching solution that uses statistical pattern matching techniques. The two-tier speech retrieval solution proposed in this paper, named Audio Search, is composed of a fast searching component followed by an ends-free Viterbi keyword spotter combined with a histogram-based phoneme duration model. Its performances have been evaluated on the 1997 HUB4 database for Chinese broadcast news. In comparison with the results of large vocabulary automatic speech recognition based systems, the results we obtained with the Audio Search system show significant improvements for both long queries (typically 5 to 7 syllables) and out-of-vocabulary (OOV) queries.

关键词： ASR, Ends-free Viterbi algorithm

来源：评论

学校读者我要写书评

暂无评论

没有更多数据了...

全选清除本页清除全部题录导出标记到“检索档案”

共118页 << < 109 110 111 112 113 114 115 116 117 118 > >>

检索报告对象比较合并检索0

隐藏清空

合并搜索

回到顶部

执行限定条件

内容：

评分：

请选择保存的检索档案：

请选择收藏分类：

订阅名称：

通借通还

温馨提示：

图书名称：

借书校区：

取书校区：

手机号码：

邮箱地址：

一卡通帐号：

电话和邮箱必须正确填写，我们会与您联系确认。

联系人：

所在院系：

联系邮箱：

联系电话：

内蒙古自治区呼和浩特市赛罕区大学西街235号邮编: 010021

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：