Continuous Multi-Band Speech Recognition using Bayesian Networks

Khalid Daoudi and Dominique Fohr and Christophe Antoine. ( 2001 )
in: IEEE Automatic Speech Recognition and Understanding Workshop - ASRU'2001, IEEE, pages 4 p

Abstract

Using the Bayesian networks framework, we present a new multi-band approach for continuous speech recognition. This new approach has the advantage to overcome all the limitations of the standard multi-band techniques. Moreover, it leads to a higher fidelity speech modeling than HMMs. We provide a preliminary evaluation of the performance of our new approach on a connected digits recognition task.

Download / Links

BibTeX Reference

@inproceedings{daoudi:inria-00107521,
 abstract = {Using the Bayesian networks framework, we present a new multi-band approach for continuous speech recognition. This new approach has the advantage to overcome all the limitations of the standard multi-band techniques. Moreover, it leads to a higher fidelity speech modeling than HMMs. We provide a preliminary evaluation of the performance of our new approach on a connected digits recognition task.},
 address = {Trento, Italy},
 author = {Daoudi, Khalid and Fohr, Dominique and Antoine, Christophe},
 booktitle = {{IEEE Automatic Speech Recognition and Understanding Workshop - ASRU'2001}},
 hal_id = {inria-00107521},
 hal_local_reference = {A01-R-226 || daoudi01b},
 hal_version = {v1},
 keywords = {speech recognition ; r{\'e}seaux bay{\'e}siens ; bayesian networks ; reocnnaissance la parole},
 month = {December},
 note = {Colloque avec actes et comit{\'e} de lecture. internationale.},
 organization = {{IEEE}},
 pages = {4 p},
 pdf = {https://hal.inria.fr/inria-00107521/file/A01-R-226.pdf},
 title = {{Continuous Multi-Band Speech Recognition using Bayesian Networks}},
 url = {https://hal.inria.fr/inria-00107521},
 year = {2001}
}