Getting More Mileage from Web Text Sources for Conversational Speech Language Modeling using Class-Dependent Mixtures

Sources of training data suitable for language modeling of conversational speech are limited. In this paper, we show how training data can be supplemented with text from the web filtered to match the style and/or topic of the target recognition task, but also that it is possible to get bigger performance gains from the data by using class-dependent interpolation of N-grams.

• Publication year

  2003

• Venue

  North American Chapter of the Association for Computational Linguistics

• Publication date

  2003-05-27

• Fields of study

  Linguistics, Computer Science

• Identifiers
  DOI 10.3115/1073483.1073486
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Text normalization with varied data sources for conversational speech language modeling

  2002cited by this paper
• Improving trigram language modeling with the World Wide Web

  2001cited by this paper
• The Meeting Project at ICSI

  2001cited by this paper
• Normalization of non-standard words

  2001cited by this paper
• THE SRI MARCH 2000 HUB-5 CONVERSATIONAL SPEECH TRANSCRIPTION SYSTEM

  2000cited by this paper
• Entropy-based Pruning of Backoff Language Models

  2000cited by this paper
• Selecting articles from the language model training corpus

  2000cited by this paper
• Enriching the Knowledge Sources Used in a Maximum Entropy Part-of-Speech Tagger

  2000cited by this paper
• Improved topic-dependent language modeling using information retrieval techniques

  1999cited by this paper
• Relevance weighting for combining multi-domain data for n-gram language modeling

  1999cited by this paper
• Exploiting both local and global constraints for multi-span statistical language modeling

  1998cited by this paper
• Just-in-time language modelling

  1998cited by this paper
• Language model adaptation using mixtures and an exponentially decaying cache

  1997cited by this paper
• Adaptive topic - dependent language modelling using word - based varigrams

  1997cited by this paper
• Transforming out-of-domain estimates to improve in-domain language models

  1997cited by this paper
• A class based approach to domain adaptation and constraint integration for empirical m-gram models

  1997cited by this paper
• A Maximum Entropy Model for Part-Of-Speech Tagging

  1996cited by this paper
• Language Modeling with Limited Domain Data

  1995cited by this paper
• SWITCHBOARD: telephone speech corpus for research and development

  1992influential reference

• Uneven success: automatic speech recognition and ethnicity-related dialects

  2022cites this paper
• Wisdom of Crowds を用いた音声言語理解の精度向上

  2020influential citation
• Improving Spoken Language Understanding by Wisdom of Crowds

  2020cites this paper
• CHAPTER 3 N-gram Language Models

  2020cites this paper
• State-of-the-Art Overview

  2019cites this paper
• Natural Language Processing Approaches in Bioinformatics

  2019cites this paper
• Session-level Language Modeling for Conversational Speech

  2018influential citation
• Articulatory Information and Multiview Features for Large Vocabulary Continuous Speech Recognition

  2018cites this paper
• The Microsoft 2017 Conversational Speech Recognition System

  2017cites this paper
• Low-Rank RNN Adaptation for Context-Aware Language Modeling

  2017cites this paper
• Two Efficient Lattice Rescoring Methods Using Recurrent Neural Network Language Models

  2016cites this paper
• Multi-Language Neural Network Language Models

  2016cites this paper
• BASED DATA AUGMENTATION FOR CANTONESE KEYWORD SPOTTING

  2016cites this paper
• Exploiting noisy web data by OOV ranking for low-resource keyword search

  2016cites this paper
• Getting more from automatic transcripts for semi-supervised language modeling

  2016cites this paper
• Generalizing and Hybridizing Count-based and Neural Language Models

  2016cites this paper
• Machine translation based data augmentation for Cantonese keyword spotting

  2016cites this paper
• Babler - Data Collection from the Web to Support Speech Recognition and Keyword Search

  2016cites this paper
• Metadiscourse tagging in academic lectures

  2016influential citation
• Efficient Training and Evaluation of Recurrent Neural Network Language Models for Automatic Speech Recognition

  2016cites this paper
• Token-level interpolation for class-based language models

  2015cites this paper
• From Language to the Real World: Entity-Driven Text Analytics

  2015cites this paper
• Paraphrastic recurrent neural network language models

  2015cites this paper
• Improving speech recognition and keyword search for low resource languages using web data

  2015cites this paper
• Data Selection With Fewer Words

  2015cites this paper
• Leveraging Twitter for Low-Resource Conversational Speech Language Modeling

  2015cites this paper
• Recurrent neural network language model training with noise contrastive estimation for speech recognition

  2015cites this paper
• Scalable Recurrent Neural Network Language Models for Speech Recognition

  2015cites this paper
• Discriminative training of context-dependent language model scaling factors and interpolation weights

  2015cites this paper
• Enhancing low resource keyword spotting with automatically retrieved web documents

  2015cites this paper
• Improving the training and evaluation efficiency of recurrent neural network language models

  2015cites this paper
• Data Selection for Statistical Machine Translation

  2014cites this paper
• Company Mention Detection for Large Scale Text Mining

  2014cites this paper
• Open-domain Language Model Construction for Speech Driven Question Answering Employing Expansion with Similar Nouns

  2014cites this paper
• Spoken Dialogue System for Information Navigation based on Statistical Learning of Semantic and Dialogue Structure

  2014influential citation
• Incorporating Weak Statistics for Low-Resource Language Modeling

  2014cites this paper
• Efficient lattice rescoring using recurrent neural network language models

  2014cites this paper
• Rapid Generation of Pronunciation Dictionaries for new Domains and Languages

  2014cites this paper
• Biber Redux: Reconsidering Dimensions of Variation in American English

  2014cites this paper
• Named Entity Recognition from Financial Press Releases

  2014cites this paper
• Abin-based ontological framework for low-resourcen-gram smoothing in language modelling

  2014cites this paper
• Web-based possibilistic language models for automatic speech recognition

  2014cites this paper
• Paraphrastic neural network language models

  2014cites this paper
• Paraphrastic language models

  2014cites this paper
• Efficient GPU-based training of recurrent neural network language models using spliced sentence bunch

  2014cites this paper
• Incorporating semantic information to selection of web texts for language model of spoken dialogue system

  2013cites this paper
• Syllable language models for Mandarin speech recognition: exploiting character language models.

  2013cites this paper
• Language model cross adaptation for LVCSR system combination

  2013cites this paper
• Unsupervised language model adaptation for automatic speech recognition of broadcast news using web 2.0

  2013cites this paper
• An IR-Inspired Approach to Recovering Named Entity Tags in Broadcast News

  2013cites this paper
• Using web text to improve keyword spotting in speech

  2013cites this paper
• Cross-domain paraphrasing for improving language modelling using out-of-domain data

  2013cites this paper
• Use of contexts in language model interpolation and adaptation

  2013cites this paper
• Transcription of Russian conversational speech

  2012cites this paper
• Efficient Language Model Construction for Spoken Dialog Systems by Inducting Language Resources of Different Languages

  2012cites this paper
• Transcribing Meetings With the AMIDA Systems

  2012cites this paper
• Automatic transcription of academic lectures from diverse disciplines

  2012cites this paper
• Revisiting the Predictability of Language: Response Completion in Social Media

  2012cites this paper
• Training a Language Model Using Webdata for Large Vocabulary Japanese Spontaneous Speech Recognition

  2011cites this paper
• Similarity Based Language Model Construction for Voice Activated Open-Domain Question Answering

  2011cites this paper
• Integrated Online Speaker Clustering and Adaptation

  2011cites this paper
• présentée à l'Université d'Avignon et des Pays de Vaucluse pour obtenir le diplôme de DOCTORAT

  2011cites this paper
• Modèles de langage ad hoc pour la reconnaissance automatique de la parole. (Ad-hoc language models for automatic speech recognition)

  2011cites this paper
• Vocabulary and Language Model Adaptation Using just One File

  2010cites this paper
• Language model adaptation using WWW documents obtained by utterance-based queries

  2010cites this paper
• Adaptation thématique non supervisée d'un système de reconnaissance automatique de la parole. (Unsupervised topic-based adaptation of an automatic speech recognition system)

  2010cites this paper
• Vocabulary and language model adaptation using just one speech file

  2010cites this paper
• Which words are hard to recognize? Prosodic, lexical, and disfluency factors that increase speech recognition error rates

  2010cites this paper
• Cheap, Fast and Good Enough: Automatic Speech Recognition with Non-Expert Transcription

  2010cites this paper
• Web Augmentation of Language Models for Continuous Speech Recognition of SMS Text Messages

  2009cites this paper
• Unsupervised acoustic and language model training with small amounts of labelled data

  2009cites this paper
• Speaker normalisation for large vocabulary multiparty conversational speech recognition

  2009cites this paper
• Filtering web text to match target genres

  2009cites this paper
• Constraint Dependency Grammars : SuperARVs , Language Modeling , and Parsing

  2009influential citation
• A System for Simultaneous Translation of Lectures and Speeches

  2009cites this paper
• Language Modeling for limited-data domains

  2009cites this paper
• An Iterative Relative Entropy Minimization-Based Data Selection Approach for n-Gram Model Adaptation

  2009cites this paper
• Analysis of low-resource acoustic model self-training

  2009cites this paper
• Part-of-speech histograms for genre classification of text

  2009cites this paper
• Useful Transcriptions of Webcast Lectures

  2009cites this paper
• Probabilistic and possibilistic language models based on the world wide web

  2009cites this paper
• Speech Recognition in Mobile Phones

  2008cites this paper
• Rapid development of an English/Farsi speech-to-speech translation system.

  2008cites this paper
• Unsupervised versus supervised training of acoustic models

  2008influential citation
• Vocabulary Adaptation Using Contextual Information and Information Retrieval

  2008cites this paper
• PROGRESS IN MEETING RECOGNITION: THE ICSI-SRI-UW SPRING 2004 EVALUATION SYSTEM

  2008cites this paper
• Utilization of Huge Written Text Corpora for Conversational Speech Recognition

  2008cites this paper
• Speech Processing for Audio Indexing

  2008cites this paper
• Natural Language Processing and the Web

  2008cites this paper
• Online vocabulary adaptation using contextual information and information retrieval

  2008cites this paper
• Speechlinks: Robust Cross-Lingual Tactical Communication Aids

  2008cites this paper
• Topic-Specific Language Model Based on Graph Spectral Approach for Speech Recognition

  2008cites this paper
• Rapid bootstrapping of statistical spoken dialogue systems

  2008cites this paper
• The AMI System for the Transcription of Speech in Meetings

  2007cites this paper
• Web resources for language modeling in conversational speech recognition

  2007cites this paper
• Random forests and the data sparseness problem in language modeling

  2007cites this paper
• Continuous space language models

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

  2007cites this paper
• Language Model Adaptation in Machine Translation from Speech

  2007cites this paper
• Transformation sharing strategies for mllr speaker adaptation

  2007cites this paper