搜索资源列表
modelbasedonspectrumprediction
- 文章展示了基于高斯混合模型的语音频谱预测方法。频谱预测可能在传包过程中预防丢包这方面起到大作用。期望最大化算法用两倍或三倍的连续语音因素来测试模型。模型被用来设计第一，儿等指令预测量。预测表用频谱分配状态来估计并和一个简单的参考模型对比。最好的预测表得到一个平均频率扭曲值是0.46dB小于参考模型-This paper presents methods for speech spectrum prediction based on Gaussian mixture models. Spec
research
- 连续语音关键词识别系统的拒识方法研究，拒识算法的一篇不错的论文-Keywords continuous speech recognition system rejection method, rejection of a good paper algorithm
improving-efficiency
- 非特定人孤立词识别系统以及桌面上非特定人特定任务的连续语音识别系统 -A isolated-word speech recognition systemon and a continuous speech recognition systemon PC
speech
- 语音识别 汉语语音识别 连续语音识别的研究进展-Continuous Mandarin Speech Recognition Speech Recognition Speech Recognition Research
Continuous-speech-
- 语音识别中关于连续语音的应用，当说话比较长时，句子比较长是语言识别的难点，本章重点解决连续语音的识别问题-On continuous speech voice recognition applications, when the talk is longer, a longer sentence is the difficulty of speech recognition, this chapter focused on solving the problem of continuous sp
Continuous-Speech-Recognition
- 连续语音识别资料，包括各种隐性马尔科夫算法。-Continuous speech recognition, including a variety of hidden Markov algorithm.
999999
- 实现马尔科夫的语音识别功能 可以实现连续语音识别，孤立词识别-Of Markov s voice recognition capabilities can achieve continuous speech recognition, isolated word recognition
young96lvcsr
- 大规模词汇的连续语音识别论文，给出了详细的算法。-I hope it helpful to you.
eVADD-DTW-HMMn
- 端点检测与一种基于DTW与HHMM的孤立词识别与连续语音识别 -endpoint detection and DTW-based HMM and isolated word recognition and CContinuous Speech Recognition
MVoiceRecognia
- 基于马尔科夫链的连续语音识别的源代码Matlab源码--Speechh Recognition Matlab source code -- -Based on Markov chains, continuous speech recognition source code Matlab source- Speechh Recognition of Matlab the source code-
Endpointdetection
- 这是一个连续语音端点检测的matlab程序，比较简单，希望可以帮到大家-This is a continuous speech endpoint detection, the matlab program is relatively simple, I hope you can help
CDessign-Patto
- 连续语音识别的隐马尔科夫cdhmm的的程序源码.绝对正确且好用 -Program source code for the continuous speech recognition Hidden Markov cdhmm absolutely correct and easy to use
voice
- 汉语连续语音识别中上下文相关的声韵母建模.pdf-Chinese continuous speech recognition context-sensitive acoustic modeling finals. Pdf
Tone-Recognition
- 调信息在汉语语音识别中具有非常重要的意义。采用支持向量机对连续汉语连续语音进行声调识别实 验，首先采用基于Teager能量算子和过零率的两级判别策略对连续语音进行浊音段提取，然后建立了适合于支持向 量机分类模型的等维声调特征向量。使用6个二类SVM模型对非特定人汉语普通话的4种声调进行分类识别，与 BP神经网络相比，支持向量杌具有更高的识别率。-Tone is an essential component for word formation in Chinese languages
xiangduiguli
- 根据dtw算法改进的，可以识别相对连续语音的信号。-According to DTW algorithm improved, can identify the relative continuous speech signal.
Language-endpoint-detection-
- 短时能量 过零率 语音端点检测 连续语音-Short-term energy Zero crossing rate Voice activity detection
SPHINX_3-
- 大词汇量语音识别系统，可以识别连续语音，是美国卡内基梅隆开发的开源语音识别系统，该程序用C编写，模块结构非常清晰，函数接口明了-Large vocabulary speech recognition system can recognize continuous speech, Carnegie Mellon developed open source speech recognition system, the program is written in C, the module struc
Speech-Corpus
- 声语音连续语音语料库，包含用于语音识别培训和测试的数据-Voiced speech continuous speech corpus, contains the data for the voice recognition training and testing
Spectral_Entropy
- 连续语音端点检测，短时能量、过零率以及谱熵相结合，健壮性好，识别率高。阈值为自适应，可适合情况下语音端点检测。-Continuous speech endpoint detection, short-term energy, zero-crossing rate and spectral entropy, robustness, high recognition rate. The threshold adaptive, may be suitable for the case of voice
Continuous-Speech-Recognition-
- 连续语音识别中的搜索策略研究, 清华大学硕士论文, 2001-Continuous Speech Recognition search strategy research, Tsinghua University, master s thesis, 2001