Maximizing the continuity in segmentation - A new approach to model, segment and recognize speech

机译：最大化分段的连续性-一种建模，分段和识别语音的新方法

获取原文

页面导航

摘要
著录项
相似文献
相关主题

摘要

This paper presents a new approach to speech modeling and recognition. The new approach consists of a statistical model to represent up to sentence-long temporal dynamics in the training data, and an algorithm to identify the matching segments with maximum continuities between the training and testing sentences. Recognition is performed by combining the longest matching segments found from the training sentences. Because of their richer and more distinct temporal dynamics, longer speech segments as whole units can be recognized with lower error rates than shorter speech segments. Therefore basing recognition on the longest matching segments optimizes the discrimination and hence recognition of speech. The new approach has been evaluated on the TIMIT database for identifying matching speech segments. The results obtained are encouraging given the very low parametric complexity of the new model.

机译：本文提出了一种新的语音建模和识别方法。新方法包括一个统计模型，该模型可以在训练数据中表示句子中最长的时间动态，以及一种算法，该算法可以识别出训练和测试句子之间具有最大连续性的匹配句段。通过组合从训练句子中找到的最长匹配段来执行识别。由于它们具有更丰富，更独特的时间动态特性，因此与较短的语音段相比，可以将较长的语音段作为整体单元以较低的错误率识别。因此，基于最长匹配段的识别可优化辨别力，从而优化语音识别。已经在TIMIT数据库上对新方法进行了评估，以识别匹配的语音片段。鉴于新模型的参数复杂度非常低，获得的结果令人鼓舞。

著录项

来源
《IEEE International Conference on Acoustics, Speech and Signal Processing;ICASSP 2009》|2009年|3849-3852|共4页
会议地点 Taipei(CT);Taipei(CT)
作者
Ji Ming;
展开▼
作者单位

Inst. of Electron. Commun. & Inf. Technol. Queen's Univ. Belfast Belfast;

展开▼
会议组织
原文格式 PDF
正文语种
中图分类
关键词
speech recognition; statistical analysis; matching segments; parametric complexity; segmentation continuity; sentence long temporal dynamics; speech segmentation; statistical model; speech modeling; temporal dynamics;

机译：语音识别;统计分析;匹配段；参数复杂度；分割连续性；句子长时态动力学；语音分割统计模型；语音建模；时间动力;

相似文献

外文文献
中文文献
专利

1. A model distance maximizing framework for speech recognizer-based speech enhancement [J] . Babaali B., Sameti H., Falk T.H. AEU: Archiv fur Elektronik und Ubertragungstechnik: Electronic and Communication . 2011,第2期

机译：基于语音识别器的语音增强的模型距离最大化框架
2. Maximizing embedding capacity for speech steganography: a segment-growing approach [J] . Baziyad Mohammed, Shahin Ismail, Rabie Tamer, Multimedia Tools and Applications . 2021,第16期

机译：最大化嵌入式言语隐写术的能力：一种成长的分割方法
3. A Constraint-Based Evolutionary Learning Approach to the Expectation Maximization for Optimal Estimation of the Hidden Markov Model for Speech Signal Modeling [J] . Huda S., Yearwood J., Togneri R. IEEE transactions on systems, man, and cybernetics. Part B, Cybernetics . 2009,第1期

机译：一种基于约束的进化学习方法，用于期望最大化，用于语音信号建模的隐马尔可夫模型的最优估计
4. MAXIMIZING THE CONTINUITY IN SEGMENTATION - A NEW APPROACH TO MODEL, SEGMENT AND RECOGNIZE SPEECH [C] . Ji Ming IEEE International Conference on Acoustics, Speech, and Signal Processing . 2009

机译：最大化分割中的连续性 - 模拟，段和识别语音的新方法
5. A maximum likelihood continuity mapping approach to discovering the intrinsic structure of speech articulation manifolds. [D] . Valdez, Patrick F. 2005

机译：一种最大似然连续性映射方法，用于发现语音发音歧管的固有结构。
6. Image segmentation for automatic particle identification in electron micrographs based on hidden Markov random field models and expectation maximization [O] . Vivek Singh, Dan C. Marinescu, Timothy S. Baker -1

机译：基于隐马尔可夫随机场模型和期望最大化的电子显微图像中颗粒自动识别的图像分割
7. A Sinusoidal Model Approach to Acoustic Landmark Detection and Segmentation for Robust Segment-Based Speech Recognition [O] . Tara N. Sainath, Timothy J. Hazen 2006

机译：基于稳健的基于语音的语音识别的声音正弦检测和分割的正弦模型方法

Maximizing the continuity in segmentation - A new approach to model, segment and recognize speech

摘要

著录项

相似文献

相关主题

期刊订阅