The 2010 CMU GALE speech-to-text system

This paper describes the latest Speech-to-Text system developed for the Global Autonomous Language Exploitation (“GALE”) domain by Carnegie Mellon University (CMU). This systems uses discriminative training, bottle-neck features and other techniques that were not used in previous versions of our system, and is trained on 1150 hours of data from a variety of Arabic speech sources. In this paper, we show how different lexica, pre-processing, and system combination techniques can be used to improve the final output, and provide analysis of the improvements achieved by the individual techniques. Index Terms: speech recognition, discriminative training, bottle-neck features

• Publication year

  2010

• Venue

  Interspeech

• Publication date

  Unknown publication date

• Fields of study

  Linguistics, Computer Science

• Identifiers
  DOI 10.21437/Interspeech.2010-439
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• The IBM 2011 GALE Arabic speech transcription system

  2011cited by this paper
• A comparative study on system combination schemes for LVCSR

  2010cited by this paper
• Recent improvements to the Cambridge Arabic Speech-to-Text systems

  2010cited by this paper
• Training and adapting MLP features for Arabic speech recognition

  2009cited by this paper
• Lexical and phonetic modeling for Arabic automatic speech recognition

  2009cited by this paper
• Generalized discriminative feature transformation for speech recognition

  2009cited by this paper
• Correlated Bigram LSA for Unsupervised Language Model Adaptation

  2008cited by this paper
• Optimizing bottle-neck features for lvcsr

  2008cited by this paper
• Boosted MMI for model and feature-space discriminative training

  2008cited by this paper
• Development of the SRI/nightingale Arabic ASR system

  2008cited by this paper
• Transcribing broadcast data using MLP features

  2008cited by this paper
• Advances in the CMU/Interact Arabic GALE Transcription System

  2007cited by this paper
• Issues in Arabic Orthography and Morphology Analysis

  2004influential reference
• Speaker segmentation and clustering in meetings

  2004cited by this paper
• SRILM - an extensible language modeling toolkit

  2002cited by this paper
• A one-pass decoder based on polymorphic linguistic context assignment

  2001cited by this paper
• Tandem connectionist feature extraction for conventional HMM systems

  2000cited by this paper
• Semi-tied covariance matrices for hidden Markov models

  1999cited by this paper
• Speaker normalization based on frequency warping

  1997cited by this paper

• Cross-Lingual Bridges with Models of Lexical Borrowing

  2016cites this paper
• Proposal : Cross-Lingual Transfer of Linguistic and Metalinguistic Knowledge via Lexical Borrowing

  2015cites this paper
• Constraint-Based Models of Lexical Borrowing

  2015cites this paper
• Automatic Speech Recognition for Low-resource Languages and Accents Using Multilingual and Crosslingual Information

  2014cites this paper
• Multilingual multilayer perceptron for rapid language adaptation between and across language families

  2013cites this paper
• Lattice-based training of bottleneck feature extraction neural networks

  2013cites this paper
• Towards single pass discriminative training for speech recognition

  2012cites this paper
• Active learning for accent adaptation in Automatic Speech Recognition

  2012cites this paper
• An Investigation on Initialization Schemes for Multilayer Perceptron Training Using Multilingual Dat

  2012cites this paper
• Enhanced Polyphone Decision Tree Adaptation for Accented Speech Recognition

  2012cites this paper
• Generalized discriminative training for speech recognition

  2012cites this paper
• Multilingual bottle-neck features and its application for under-resourced languages

  2012cites this paper
• Semi-supervised learning for speech recognition in the context of accent adaptation

  2012cites this paper
• Acoustic and Lexical Modeling Techniques for Accented Speech Recognition

  2012cites this paper
• Temporal Patterns ( TRAPs ) in Janus Recognition Toolkit Student Project of Tatiana Glushkova

  2012cites this paper
• Temporal Patterns (TRAPs) in Janus Recognition Toolkit

  2012cites this paper
• A Study on Speaker Normalized MLP Features in LVCSR

  2011cites this paper
• The 2011 KIT QUAERO speech-to-text system for Spanish

  2011cites this paper
• Analysis of Dialectal Influence in Pan-Arabic ASR

  2011cites this paper
• Generalized Baum-Welch Algorithm and its Implication to a New Extended Baum-Welch Algorithm

  2011cites this paper