Local Prior Matching (LPM)

Local Prior Matching is a semi-supervised objective for speech recognition that distills knowledge from a strong prior (e.g. a language model) to provide learning signal to a discriminative model trained on unlabeled speech. The LPM objective minimizes the cross entropy between the local prior and the model distribution, and is minimized when $q_{y\mid{x}} = \bar{p}_{y\mid{x}}$. Intuitively, LPM encourages the ASR model to assign posterior probabilities proportional to the linguistic probabilities of the proposed hypotheses.

相关学科： Face Image RetrievalGraph SelectionLearning with noisy labelsTable SearchFPMSequential Pattern MiningTerm ExtractionMusic CompositionSCNModel Discovery

学科讨论

暂无讨论内容，你可以

学科管理组

暂无学科课代表，你可以申请成为课代表

重要学者

Jie Zhang

193659 被引用，5433 篇论文

Niels Birbaumer

88695 被引用，932 篇论文

Wil M. P. van der Aalst

35971 被引用，756 篇论文

Stan Z. Li

34357 被引用，557 篇论文

Michael L. Littman

29776 被引用，342 篇论文

Lawrence Carin

25274 被引用，999 篇论文

Tom Walsh

21672 被引用，250 篇论文

Marlon Dumas

20819 被引用，492 篇论文

Kathleen M. Carley

19421 被引用，649 篇论文

Jonathan S. Turner

16668 被引用，215 篇论文