Method for aligning text with audio signals
作者： Oren Glickman , Christopher Frank Joerg
DOI:
关键词:
摘要: In a computerized method, text segments of file are aligned with audio an file. The includes written words, and the spoken words. A vocabulary language model generated from segment. word list is recognized segment using model. segment, corresponding anchors chosen in Using anchors, partitioned into unaligned according to anchors. These steps repeated for any until termination condition reached.
lens.org 参考文章(5)
Michael Alan Picheny, Dimitri Kanevsky, David Nahamoo, Hamed A. Ellozy, Michelle Y. Kim, Wlodek Wlodzimierz Zadrozny, Automatic indexing and aligning of audio and text using speech recognition ,(1995)
Raj Reddy, Kai-Fu Lee, Large-vocabulary speaker-independent continuous speech recognition: the sphinx system Carnegie Mellon University. ,(1988)
Charles T. Hemphill, Barbara J. Wheatley, Thomas D. Fisher, George R. Doddington, System and method for time aligning speech ,(1992)
A.G. Hauptmann, H.D. Wactlar, Indexing and search of multimodal information international conference on acoustics, speech, and signal processing. pp. 195- 198 ,(1997) , 10.1109/ICASSP.1997.599599