On condition of pure sound,this paper put forward a sensitive speech extraction based on short-time Fourier transform,which used three FIR digital filter extract sensitive speech in different frequency range.

Bimodal speech extraction method based on particle filtering;

The present study examined the relationship between phonological sensitivity and the memory of Chinese characters and English words with Chinese Mainland Putonghua-speaking children as subjects.

In order to express dynamic characteristic for speech,feature difference and multi-frames speech feature extraction based on association relationship of lip frames are introduced.

Wavelet transform is very effective for picking up the pitch detection from a single source,but there is no effective way to obtain the pitch information from the pitch detection in overlapping speech with multi-sources.

The continuous Chinese tone of three syllabic word is very complicated and its pitch detection is more difficult than that of isolated word because of the effects of vocal response,syllable conjunction,coarticulation and tone sandhi.

An approach of the pitch detection for obtaining speech tone of Chinese trisyllabic words on the basis of the wavelet transforms is proposed .