• No results found

word n-gram models

Using sub word n gram models for dealing with OOV in large vocabulary speech recognition for Latvian

Using sub word n gram models for dealing with OOV in large vocabulary speech recognition for Latvian

... Intuitively, any OOV improvement should also result in improvement of recognition quality. For example, the same 200K baseline system shows about 27% WER on a subset of evaluation data with no OOV words. However, ...

5

Reduced n gram Models for English and Chinese Corpora

Reduced n gram Models for English and Chinese Corpora

... of n-grams. To avoid these problems, the reduced n-grams’ approach previously developed by O’Boyle (1993) can be ...reduced n-gram language model can store an entire corpus’s phrase-history ...

7

N gram Counts and Language Models from the Common Crawl

N gram Counts and Language Models from the Common Crawl

... sequence models built on the in-domain subset of the parallel corpus using Kneser-Ney smoothed 7-gram models and as additional factors in phrase translation mod- els (Koehn and Hoang, ...guage ...

6

Letter N Gram based Input Encoding for Continuous Space Language Models

Letter N Gram based Input Encoding for Continuous Space Language Models

... language models uses a 1-of-n coding to insert a word from the vocabulary into the ...the word in the vocabulary is set to one and the rest to ...an n-gram Boltzmann machine ...

10

The RWTH Aachen University English Romanian Machine Translation System for WMT 2016

The RWTH Aachen University English Romanian Machine Translation System for WMT 2016

... following models: Phrase translation probabilities and lexical smoothing in both directions, word and phrase penalty, distance- based reordering model, n -gram target language models ...

6

HIVEC: A Hierarchical Approach for Vector Representation Learning of Graphs

HIVEC: A Hierarchical Approach for Vector Representation Learning of Graphs

... as word embed- dings. These neural language models improve classic n-gram language models by using continuous vector representations for ...traditional n-gram ...

7

Naive Bayes and BiLSTM Ensemble for Discriminating between Mainland and Taiwan Variation of Mandarin Chinese

Naive Bayes and BiLSTM Ensemble for Discriminating between Mainland and Taiwan Variation of Mandarin Chinese

... the n-gram features based models and the ensemble ...ing models, we implement them using Keras 2 library with Tensorflow ...and word em- beddings have 300 dimensions. Word ...

8

Self Organizing n gram Model for Automatic Word Spacing

Self Organizing n gram Model for Automatic Word Spacing

... One of the most simple and strong models for automatic word spacing is -gram model. In spite of the advantages of the -gram model, its prob- lem should be also considered for achieving high ...

8

Better Word Embeddings by Disentangling Contextual n Gram Information

Better Word Embeddings by Disentangling Contextual n Gram Information

... Pre-trained word vectors are ubiquitous in Natural Language Processing ...training word em- beddings jointly with bigram and even trigram embeddings, results in improved unigram em- ...training word ...

7

Character n-Gram Embeddings to Improve RNN Language Models

Character n-Gram Embeddings to Improve RNN Language Models

... character n-grams based on research in the field of word embedding construction (Wieting et ...constructs word embeddings from character n- gram embeddings and combines them with ...

9

Faster and Smaller N Gram Language Models

Faster and Smaller N Gram Language Models

... the word in the node, and one for either a pointer to the par- ent of the node or a list of pointers to ...the word, a 64-bit memory 2 pointer to the list of children, and a 32-bit floating point num- ber ...

10

Subsegmental language detection in Celtic language text

Subsegmental language detection in Celtic language text

... per word level, yet we would like to include supervised methods and features talked about in this research to improve our efficiency while dealing with ...character n-gram models for backoff, ...

5

Hierarchical Bayesian Language Modelling for the Linguistically Informed

Hierarchical Bayesian Language Modelling for the Linguistically Informed

... language models in machine translation (MT) and automatic speech recognition (ASR) is widely ...recognised. n-gram models, in particular ones using Kneser-Ney (KN) smoothing, have become the ...

10

A Challenge Set for Advancing Language Modeling

A Challenge Set for Advancing Language Modeling

... 4-gram models built with the CMU toolkit achieved 36 to 39 ...the word scores at specific positions relative to the focus ...following word, and combin- ing it with the score of the ...

8

Statistical Input Method based on a Phrase Class n gram Model

Statistical Input Method based on a Phrase Class n gram Model

... An n-gram model is generally used for many ...bi-gram models are often used. However, bi-gram models can not refer to a long ...tri-gram models is larger than that ...

14

Predicting Sentences using N Gram Language Models

Predicting Sentences using N Gram Language Models

... In the context of natural language, several typ- ing assistance tools for apraxic (Garay-Vitoria and Abascal, 2004; Zagler and Beck, 2002) and dyslexic (Magnuson and Hunnicutt, 2002) persons have been developed. These ...

8

Investigating Speech Recognition for Improving Predictive AAC

Investigating Speech Recognition for Improving Predictive AAC

... or word predictions can help accelerate the communication of users of high-tech AAC ...tion word error rates of 7–16%, our ensemble of N-gram and recurrent neural network lan- guage ...

7

Distributed representation and estimation of WFST based n gram models

Distributed representation and estimation of WFST based n gram models

... Billion Word Benchmark (BWB) cor- pus (Chelba et ...5- gram language models with different parameteri- zations for determining the model ...billion word collection of search queries (SQ), also ...

10

Language Identification of Short Text Segments with N-gram Models

Language Identification of Short Text Segments with N-gram Models

... with n- gram language models. Because of the compact models that do not need word-based features, this approach is well suited for language identification tasks that have dozens of ...

8

Detecting Traffic Information From Social Media Texts With Deep Learning Approaches

Detecting Traffic Information From Social Media Texts With Deep Learning Approaches

... bag-of- word model to learn word embedding representations based on a data set of three billion ...words, word embedding can capture semantic similarity between words and has been proved effective in ...

10

Show all 10000 documents...

Related subjects