Abstract | ||
---|---|---|
This paper presents a language model adaptation technique to build a single static language model from a set of language models each trained on a separate text corpus while aiming to maximize the likelihood of an adaptation data set given as a development set of sentences. The proposed model can be considered as a mixture of mixture language models. The mixture model at the top level is a sentence-level mixture model where each sentence is assumed to be drawn from one of a discrete set of topic or task clusters. After selecting a cluster, each n-gram is assumed to be drawn from one of the given n-gram language models. We estimate cluster mixture weights and n-gram language model mixture weights for each cluster using the expectation-maximization (EM) algorithm to seek the parameter estimates maximizing the likelihood of the development sentences. This mixture of mixture models can be represented efficiently as a static n-gram language model using the previously proposed Bayesian language model interpolation technique. We show a significant improvement with this technique (both perplexity and WER) compared to the standard one level interpolation scheme. |
Year | DOI | Venue |
---|---|---|
2013 | 10.1109/ASRU.2013.6707701 | 2013 IEEE WORKSHOP ON AUTOMATIC SPEECH RECOGNITION AND UNDERSTANDING (ASRU) |
Keywords | Field | DocType |
language model, adaptation, interpolation, mixture models, bayesian, speech recognition | Perplexity,Pattern recognition,Computer science,Interpolation,Text corpus,Speech recognition,n-gram,Artificial intelligence,Sentence,Mixture model,Language model,Bayesian probability | Conference |
Citations | PageRank | References |
0 | 0.34 | 6 |
Authors | ||
4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Hasim Sak | 1 | 690 | 39.56 |
Cyril Allauzen | 2 | 690 | 47.64 |
Kaisuke Nakajima | 3 | 8 | 1.34 |
Françoise Beaufays | 4 | 341 | 27.76 |