DE eng

Search in the Catalogues and Directories

Hits 1 – 10 of 10

1
Maximum Entropy Language Modeling with Non-Local and Syntactic Dependencies
In: http://www.cs.jhu.edu/~junwu/gbo.ps (2002)
BASE
Show details
2
Smoothing Issues in the Structured Language Model
In: http://cs.jhu.edu/~junwu/eurospeech01.ps (2001)
BASE
Show details
3
Smoothing Issues in the Structured Language Mode
In: http://www.clsp.jhu.edu/~woosung/pdf/euro01.pdf (2001)
BASE
Show details
4
Syntactic Heads In Statistical Language Modeling
In: http://www.cs.jhu.edu/~junwu/icassp2000.ps (2000)
BASE
Show details
5
Combining Nonlocal, Syntactic And N-Gram Dependencies In Language Modeling
In: http://www.cs.jhu.edu/~junwu/eurospeech.ps (1999)
BASE
Show details
6
A Maximum Entropy Language Model Integrating N-Grams And Topic Dependencies For Conversational Speech Recognition
In: http://www.cs.jhu.edu/~junwu/topic-lm.ps (1999)
BASE
Show details
7
A maximum entropy language model integrating ngrams and topic dependencies for conversational speech reconition
In: http://www.mirlab.org/conference_papers/International_Conference/ICASSP 1999/PDF/AUTHOR/IC992192.PDF (1999)
BASE
Show details
8
A Maximum Entropy Language Model with Topic Sensitive Features
In: http://www.cs.jhu.edu/~junwu/memodel.ps (1998)
BASE
Show details
9
Building A Topic-Dependent Maximum Entropy Model For Very Large Corpora
In: http://www.cs.jhu.edu/~junwu/icassp02.ps (1217)
Abstract: Maximum entropy (ME) techniques have been successfully used to combine different sources of linguistically meaningful constraints in language models. However, most of the current ME models can only be used for small corpora, since the computational load in training ME models for large corpora is unbearable. This problem is especially severe when non-local dependencies are considered. In this paper, we show how to train and use topic-dependent ME models efficiently for a very large corpus, Broadcast News (BN). The training time is greatly reduced by hierarchical training and divide-and-conquer approaches. The computation in using the model is also simplified by pre-normalizing the denominators of the ME model. We report new speech recognition results showing improvement with the topic model relative to the standard N-gram model for the Broadcast News task.
URL: http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.24.9889
http://www.cs.jhu.edu/~junwu/icassp02.ps
BASE
Hide details
10
Maximum Entropy Techniques for Exploiting Syntactic, Semantic and Collocational Dependencies in Language Modeling
In: http://www.cs.jhu.edu/~junwu/cslpaper.ps
BASE
Show details

Catalogues
0
0
0
0
0
0
0
Bibliographies
0
0
0
0
0
0
0
0
0
Linked Open Data catalogues
0
Online resources
0
0
0
0
Open access documents
10
0
0
0
0
© 2013 - 2024 Lin|gu|is|tik | Imprint | Privacy Policy | Datenschutzeinstellungen ändern