Word segmentation is a decisive part in any speech to text conversion. Works have been done on popular languages for word boundary, especially on English. But a very few work is conducted on Bengali language, especially on colloquial speech. In speech recognition word boundary detection is significant and crucial task. To understand and fixing the problems on efficiently detecting where the words are present in a signal is still challenging. In this paper, we extract the feature of the existence of words based on the pitch profile of a speech. We cluster the pitch profile and Silhouette index is used to select the words boundary with the help of Convolutional Neural Network (CNN)..
Alan : Eğitim Bilimleri
Dergi Türü : Uluslararası
Benzer Makaleler | Yazar | # |
---|
Makale | Yazar | # |
---|