Zobrazeno 1 - 10
of 31
pro vyhledávání: '"Enrico Bocchieri"'
Publikováno v:
IEEE/ACM Transactions on Audio, Speech, and Language Processing. 23:1348-1357
Most of the state-of-the-art ASR systems take as input a single type of acoustic features, dominated by the traditional feature schemes, i.e., MFCCs or PLPs. However, these features cannot model rapid, intra-frame phenomena present in the actual spee
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::b1b42a061f1ea48586cb588f069f7598
https://doi.org/10.1109/taslp.2015.2430815
https://doi.org/10.1109/taslp.2015.2430815
Autor:
B. Kan-Wing Mak, Enrico Bocchieri
Publikováno v:
IEEE Transactions on Speech and Audio Processing. 9:378-387
It generally takes a long time and requires a large amount of speech data to train hidden Markov models for a speech recognition task of a reasonably large vocabulary. Previously, we proposed a compact acoustic model called "subspace distribution clu
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::d84503c6a7454696c7e05ccf471f263a
https://doi.org/10.1109/89.917683
https://doi.org/10.1109/89.917683
Autor:
Enrico Bocchieri, Brian Mak
Publikováno v:
IEEE Transactions on Speech and Audio Processing. 9:264-275
Most contemporary laboratory recognizers require too much memory to run, and are too slow for mass applications. One major cause of the problem is the large parameter space of their acoustic models. In this paper, we propose a new acoustic modeling m
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::3971cb13fe627e54e6c5236ced241062
https://doi.org/10.1109/89.906000
https://doi.org/10.1109/89.906000
Publikováno v:
INTERSPEECH
In the current ASR systems the presence of competing speakers greatly degrades the recognition performance. This phenomenon is getting even more prominent in the case of hands-free, far-field ASR systems like the “Smart-TV” systems, where reverbe
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::73a4d43d38440c52baf127419966fa66
https://doi.org/10.21437/interspeech.2013-281
https://doi.org/10.21437/interspeech.2013-281
Publikováno v:
ICASSP
Micro-modulation components such as the formant frequencies are very important characteristics of spoken speech that have allowed great performance improvements in small-vocabulary ASR tasks. Yet they have limited use in large vocabulary ASR applicat
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::4fcab7ff59bc4b9861c1fae22076528b
https://doi.org/10.1109/icassp.2013.6638960
https://doi.org/10.1109/icassp.2013.6638960
Autor:
Michael S. Phillips, Enrico Bocchieri, Iker Arizmendi, Chao Wang, Jay G. Wilpon, Mazin Gilbert, Diamantino Caseiro, Andrej Ljolje, Vincent Goffin
Publikováno v:
INTERSPEECH
A Mobile Virtual Assistant (MVA) is a communication agent that recognizes and understands free speech, and performs actions such as retrieving information and completing transactions. One essential characteristic of MVAs is their ability to learn and
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::3d2116773944ca006175e9f38778e7e8
https://doi.org/10.21437/interspeech.2011-416
https://doi.org/10.21437/interspeech.2011-416
Publikováno v:
ICASSP
In previously published work, we have proposed a novel feature extraction algorithm, based on the Teager-Kaiser energy estimates, that approximates human auditory characteristics and that is more robust to sub-band noise than the mean-square estimate
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::13b9580b8ec0e67a4e03277e92f0608a
https://doi.org/10.1109/icassp.2011.5947351
https://doi.org/10.1109/icassp.2011.5947351
Publikováno v:
ICASSP
This paper reports on the development and advances in automatic speech recognition for the AT&T Speak4it® voice-search application. With Speak4it as real-life example, we show the effectiveness of acoustic model (AM) and language model (LM) estimati
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::bab4c4b03827477e7274932334f1fa05
https://doi.org/10.1109/icassp.2011.5947451
https://doi.org/10.1109/icassp.2011.5947451
Publikováno v:
International Journal of Pattern Recognition and Artificial Intelligence. :783-799
Recently, much interest has been generated regarding speech recognition systems based on Hidden Markov Models (HMMs) and neural network (NN) hybrids. Such systems attempt to combine the best features of both models: the temporal structure of HMMs and
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::d1c534c8ee342b8822e756db627f1501
https://doi.org/10.1142/s021800149300039x
https://doi.org/10.1142/s021800149300039x
Autor:
Enrico Bocchieri, Jay G. Wilpon
Publikováno v:
Computer Speech & Language. 7:229-246
Over the last several years, a major factor in reducing the error rate on most speech recognition systems has been the addition of new feature components to the frame vectors. However, because of the larger dimensionality of the frame feature vector,
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::fc09294868b4c81ad3a9b281307ca7dc
https://doi.org/10.1006/csla.1993.1012
https://doi.org/10.1006/csla.1993.1012