搜索资源列表
dtwpaper
- 采用DTW（动态时间规整）的方法，实现简单语音（0-9的数字）的识别-DTW algorithm in isolated word speech recognition when the focus on the time warping and voice measure of calculation, but not on the reliability and validity of data for analysis. This paper presents an improved en
Mel_RBF
- 基于Mel倒谱特征和RBF网络的孤立词语音识别方法-Mel-based cepstral features and RBF networks isolated word speech recognition method
softwarecode
- 中文分词是中文信息处理中的重要环节，中文分词技术广泛应用于自动翻译、文本检索、语音识别、文本校对、人工智能以及搜索引擎技术等领域。中文分词算法的选择，中文词库的构建方式，词库中词条的完备性在很大程度上与中文分词系统性能紧密相关。-Chinese word segmentation in Chinese information processing is an important part of Chinese word segmentation technology is widely used
bp
- 基于bp神经网络的孤立词识别，有训练好的鼓励词，是语音识别的基础。-Isolated word recognition based on bp neural network trained to encourage word speech recognition.
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
speechhmm
- 运用HMM进行语音识别之后的孤立词识别，里面函数俱全。可以直接运用，简单方便-Speech recognition with HMM isolated word recognition, which function and taste. Can directly use, simple and convenient
HMM-isolated-word-speech-recognition
- 一种基于隐马尔科夫模型的孤立词的的语音识别实验，可以试验0到9的数字语音识别。-An isolated word speech recognition experiment based on the hidden Markov model, can test 0 to 9 digit speech recognition.