Phrase-based language models for speech recognition

Including phrases in the vocabulary list can improve ngram language models used in speech recognition. In this paper, we report results of automatic extraction of phrases from the training text using frequency, likelihood, and correlation criteria. We show how a language model built from a vocabulary that includes useful phrases can systematically improve language model perplexity in a natural language call-routing task and the 20K-Nov92 Wall Street Journal evaluation. We also discuss the impact of such phrase-based language models on recognition word error rate.

• Publication year

  1999

• Venue

  EUROSPEECH

• Publication date

  1999-09-05

• Fields of study

  Linguistics, Computer Science

• Identifiers
  DOI 10.21437/Eurospeech.1999-404x
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Language-model optimization by mapping of corpora

  1998cited by this paper
• Language modeling for content extraction in human-computer dialogues

  1998cited by this paper
• A spoken language system for automated call routing

  1997cited by this paper
• Class phrase models for language modeling

  1996cited by this paper
• Statistical language modeling using a variable context length

  1996cited by this paper
• Phrase bigrams for continuous speech recognition

  1995cited by this paper

• Learning from past mistakes: improving automatic speech recognition output via noisy-clean phrase context modeling

  2018cites this paper
• Document Retrieval Using Entity-Based Language Models

  2016cites this paper
• Phrase-Based Language Model in Statistical Machine Translation

  2016cites this paper
• Phrase Based Language Model for Statistical Machine Translation - empirical study

  2015cites this paper
• Personalization of word-phrase-entity language models

  2015cites this paper
• Word-phrase-entity language models: getting more mileage out of n-grams

  2014cites this paper
• Lexicon optimization based on discriminative learning for automatic speech recognition of agglutinative language

  2014cites this paper
• Learning Constructions of Natural Language: Statistical Models and Evaluations

  2012cites this paper
• Generating compound words with high order n-gram information in large vocabulary speech recognition systems

  2011cites this paper
• Integrating history-length interpolation and classes in language modeling

  2011cites this paper
• Speech Recognition System of Slovenian Broadcast News

  2011cites this paper
• Lattice rescoring methods for statistical machine translation

  2010cites this paper
• Phrase classes in two-level language models for ASR

  2009cites this paper
• Phrasal Segmentation Models for Statistical Machine Translation

  2008cites this paper
• Incorporating linguistic structure into maximum entropy language models

  2008cites this paper
• Different Approaches to Class-Based Language Models Using Word Segments

  2008cites this paper
• Two Approaches to Class-Based Language Models for ASR

  2007cites this paper
• Informations morpho-syntaxiques et adaptation thématique pour améliorer la reconnaissance de la parole

  2007cites this paper
• Segment-Based Classes for Language Modeling Within the Field of CSR

  2007cites this paper
• Utilisation de la linguistique en reconnaissance de la parole : un état de l'art

  2006cites this paper
• Spoken language understanding using layered n-gram modeling

  2005cites this paper
• Integrating multiple layers of concept information into n-gram modeling for spoken language understanding

  2005cites this paper
• Korean large vocabulary continuous speech recognition with morpheme-based recognition units

  2003cites this paper
• Adaptive language models for spoken dialogue systems

  2002cites this paper
• Dynamic Topic Identification : Introduction of Trigger pairs in the Cache Model

  2002cites this paper
• A new lexicon optimization method for LVCSR based on linguistic and acoustic characteristics of words

  2002cites this paper
• Learning Strategies In A Grammar Induction Framework

  2001cites this paper
• OASIS natural language call steering trial

  2001cites this paper
• Data-driven approach to designing compound words for continuous speech recognition

  2001influential citation
• Compound splitting and lexical unit recombination for improved performance of a speech recognition system for German parliamentary speeches

  2000cites this paper
• Natural language call steering for service applications

  2000cites this paper