HMM2- A novel approach to HMM emission probability estimation. Weber, K., Bengio, S., & Bourlard, H. In Proceedings of the International Conference on Speech and Language Processing, ICSLP, Beijing, China, October, 2000.
Paper abstract bibtex In this paper, we discuss and investigate a new method to estimate local emission probabilities in the framework of hidden Markov models (HMM). Each feature vector is considered to be a sequence and is supposed to be modeled by yet another HMM. Therefore, we call this approach `HMM2'. There is a variety of possible topologies of such HMM2 systems, e.g. incorporating trellis or ergodic HMM structures. Preliminary HMM2 speech recognition experiments on cepstral and spectral features yielded worse results than state-of-the-art systems. However, we believe that HMM2 systems have a lot of potential advantages and are therefore worth investigating further.
@inproceedings{weber:2000:icslp,
author = {K. Weber and S. Bengio and H. Bourlard},
title = {{HMM2}- A novel approach to {HMM} emission probability estimation},
booktitle = {Proceedings of the International Conference on Speech and Language Processing, {ICSLP}},
year = 2000,
address = {Beijing, China},
month = {October},
url = {publications/ps/rr00-30.ps.gz},
pdf = {publications/pdf/rr00-30.pdf},
djvu = {publications/djvu/rr00-30.djvu},
original={ps/rr00-30.ps.gz},
topics = {speech},
abstract = {In this paper, we discuss and investigate a new method to estimate local emission probabilities in the framework of hidden Markov models (HMM). Each feature vector is considered to be a sequence and is supposed to be modeled by yet another HMM. Therefore, we call this approach `HMM2'. There is a variety of possible topologies of such HMM2 systems, e.g. incorporating trellis or ergodic HMM structures. Preliminary HMM2 speech recognition experiments on cepstral and spectral features yielded worse results than state-of-the-art systems. However, we believe that HMM2 systems have a lot of potential advantages and are therefore worth investigating further.},
categorie = {C}
}
Downloads: 0
{"_id":"auLzezppyovjYAs5S","bibbaseid":"weber-bengio-bourlard-hmm2anovelapproachtohmmemissionprobabilityestimation-2000","authorIDs":[],"author_short":["Weber, K.","Bengio, S.","Bourlard, H."],"bibdata":{"bibtype":"inproceedings","type":"inproceedings","author":[{"firstnames":["K."],"propositions":[],"lastnames":["Weber"],"suffixes":[]},{"firstnames":["S."],"propositions":[],"lastnames":["Bengio"],"suffixes":[]},{"firstnames":["H."],"propositions":[],"lastnames":["Bourlard"],"suffixes":[]}],"title":"HMM2- A novel approach to HMM emission probability estimation","booktitle":"Proceedings of the International Conference on Speech and Language Processing, ICSLP","year":"2000","address":"Beijing, China","month":"October","url":"publications/ps/rr00-30.ps.gz","pdf":"publications/pdf/rr00-30.pdf","djvu":"publications/djvu/rr00-30.djvu","original":"ps/rr00-30.ps.gz","topics":"speech","abstract":"In this paper, we discuss and investigate a new method to estimate local emission probabilities in the framework of hidden Markov models (HMM). Each feature vector is considered to be a sequence and is supposed to be modeled by yet another HMM. Therefore, we call this approach `HMM2'. There is a variety of possible topologies of such HMM2 systems, e.g. incorporating trellis or ergodic HMM structures. Preliminary HMM2 speech recognition experiments on cepstral and spectral features yielded worse results than state-of-the-art systems. However, we believe that HMM2 systems have a lot of potential advantages and are therefore worth investigating further.","categorie":"C","bibtex":"@inproceedings{weber:2000:icslp,\n author = {K. Weber and S. Bengio and H. Bourlard}, \n title = {{HMM2}- A novel approach to {HMM} emission probability estimation}, \n booktitle = {Proceedings of the International Conference on Speech and Language Processing, {ICSLP}},\n year = 2000,\n address = {Beijing, China},\n month = {October},\n url = {publications/ps/rr00-30.ps.gz},\n pdf = {publications/pdf/rr00-30.pdf},\n djvu = {publications/djvu/rr00-30.djvu},\n original={ps/rr00-30.ps.gz},\n topics = {speech},\n abstract = {In this paper, we discuss and investigate a new method to estimate local emission probabilities in the framework of hidden Markov models (HMM). Each feature vector is considered to be a sequence and is supposed to be modeled by yet another HMM. Therefore, we call this approach `HMM2'. There is a variety of possible topologies of such HMM2 systems, e.g. incorporating trellis or ergodic HMM structures. Preliminary HMM2 speech recognition experiments on cepstral and spectral features yielded worse results than state-of-the-art systems. However, we believe that HMM2 systems have a lot of potential advantages and are therefore worth investigating further.},\n categorie = {C}\n}\n\n","author_short":["Weber, K.","Bengio, S.","Bourlard, H."],"key":"weber:2000:icslp","id":"weber:2000:icslp","bibbaseid":"weber-bengio-bourlard-hmm2anovelapproachtohmmemissionprobabilityestimation-2000","role":"author","urls":{"Paper":"http://bengio.abracadoudou.com/publications/ps/rr00-30.ps.gz"},"downloads":0},"bibtype":"inproceedings","biburl":"http://bengio.abracadoudou.com/samy.bib","creationDate":"2020-03-18T03:43:27.470Z","downloads":0,"keywords":[],"search_terms":["hmm2","novel","approach","hmm","emission","probability","estimation","weber","bengio","bourlard"],"title":"HMM2- A novel approach to HMM emission probability estimation","year":2000,"dataSources":["9NCW2CDr4M3s5DvNX"]}