一种噪音环境下的基于特征口形的音频视频混合连续语音识别系统

An Eigen Mouth Based Audio Visual Continuous Speech Recognition System in Noisy Environments

文章抓住人类语音感知多模型的特点,尝试建立一个在噪音环境下的基于音频和视频复合特征的连续语音识别系统。在视频特征提取方面,引入了一种基于特征口形的提取方法。识别实验证明,这种视频特征提取方法比传统DCT、DWT方法能够带来更高的识别率;基于特征口形的音频-视频混合连续语音识别系统具有很好的抗噪性。

Considering that human speech perception is inherently a multi-modal process,the paper tries to develop a continuous speech recognition system based on audio-visual fusion,which is used in noisy environments.In the visual feature extraction,an eigen mouth based method is introduced.Experimental results show that the feature extraction method presented in this paper outperforms the traditional methods,such as DCT and DWT.The experiments also show that the audio-visual continuous speech recognition system is robust in noisy environments.