| 2025 | ACL | Failing Forward: Improving Generative Error Correction for ASR with Synthetic Data and Retrieval Augmentation. | Sreyan Ghosh, Mohammad Sadegh Rasooli, Michael Levit, Peidong Wang, Jian Xue, Dinesh Manocha, Jinyu Li |
| 2018 | Interspeech | What to Expect from Expected Kneser-Ney Smoothing. | Michael Levit, Sarangarajan Parthasarathy, Shuangyu Chang |
| 2017 | Interspeech | Don't Count on ASR to Transcribe for You: Breaking Bias with Two Crowds. | Michael Levit, Yan Huang, Shuangyu Chang, Yifan Gong |
| 2016 | Interspeech | Word-Phrase-Entity Recurrent Neural Networks for Language Modeling. | Michael Levit, Sarangarajan Parthasarathy, Shuangyu Chang |
| 2015 | ASRU | Discriminative training of context-dependent language model scaling factors and interpolation weights. | Shuangyu Chang, Abhik Lahiri, Issac Alphonso, Barlas Oguz, Michael Levit, Benot Dumoulin |
| 2015 | ICASSP | Token-level interpolation for class-based language models. | Michael Levit, Andreas Stolcke, Shuangyu Chang, Sarangarajan Parthasarathy |
| 2015 | Interspeech | Personalization of word-phrase-entity language models. | Michael Levit, Andreas Stolcke, R. Subba, Sarangarajan Parthasarathy, Shuangyu Chang, S. Xie, T. Anastasakos, Benot Dumoulin |
| 2014 | Interspeech | Word-phrase-entity language models: getting more mileage out of n-grams. | Michael Levit, Sarangarajan Parthasarathy, Shuangyu Chang, Andreas Stolcke, Benot Dumoulin |
| 2013 | ASRU | Accelerating recurrent neural network training via two stage classes and parallelization. | Zhiheng Huang, Geoffrey Zweig, Michael Levit, Benot Dumoulin, Barlas Oguz, Shawn Chang |
| 2013 | Interspeech | Improving unsupervised language model adaptation with discriminative data filtering. | Shuangyu Chang, Michael Levit, Partha Parthasarathy, Benot Dumoulin |
| 2012 | ICASSP | End-to-end speech recognition accuracy metric for voice-search tasks. | Michael Levit, Shuangyu Chang, Bruce Buntschuh, Nick Kibre |
| 2009 | ASRU | Garbage modeling with decoys for a sequential recognition scenario. | Michael Levit, Shuangyu Chang, Bruce Buntschuh |
| 2008 | ICASSP | An iterative unsupervised learning method for information distillation. | Kamand Kamangar, Dilek Hakkani-Tr, Gkhan Tr, Michael Levit |
| 2008 | Interspeech | Unsupervised learning of edit parameters for matching name variants. | Daniel Gillick, Dilek Hakkani-Tr, Michael Levit |
| 2007 | ASRU | Integrating several annotation layers for statistical information distillation. | Michael Levit, Dilek Hakkani-Tr, Gkhan Tr, Daniel Gillick |
| 2007 | Interspeech | Exploiting information extraction annotations for document retrieval in distillation tasks. | Dilek Hakkani-Tr, Gkhan Tr, Michael Levit |
| 2007 | Interspeech | Selecting on-topic sentences from natural language corpora. | Michael Levit, Elizabeth Boschee, Marjorie Freedman |
| 2005 | ICASSP | "Of All Things the Measure Is Man" : Automatic Classification of Emotions and Inter-Labeler Consistency. | Stefan Steidl, Michael Levit, Anton Batliner, Elmar Nth, Heinrich Niemann |
| 2004 | Interspeech | Aspects of named entity processing. | Michael Levit, Allen L. Gorin, Patrick Haffner, Hiyan Alshawi, Elmar Nth |
| 2003 | Interspeech | Context-sensitive evaluation and correction of phone recognition output. | Michael Levit, Hiyan Alshawi, Allen L. Gorin, Elmar Nth |
| 2002 | Interspeech | Using EM-trained string-edit distances for approximate matching of acoustic morphemes. | Michael Levit, Elmar Nth, Allen L. Gorin |
| 2001 | Interspeech | Multipass algorithm for acquisition of salient acoustic morphemes. | Michael Levit, Allen L. Gorin, Jeremy H. Wright |
| 1999 | ICASSP | Discriminative estimation of interpolation parameters for language model classifiers. | Volker Warnke, Stefan Harbeck, Elmar Nth, Heinrich Niemann, Michael Levit |