You are here: Home / Publications / Research reports

Research reports

  1. Dimitri Palaz, Mathew Magimai.-Doss and Ronan Collobert. End-to-End Acoustic Modeling using Convolutional Neural Networks for Automatic Speech Recognition. Idiap Research Report, Idiap-RR-18-2016, 2016. (pdf)
  2. Marzieh Razavi, Ramya Rasipuram and Mathew Magimai.-Doss. Towards Multiple Pronunciation Generation in Acoustic G2P Conversion Framework. Idiap Research Report, Idiap-RR-34-2015, 2015. (pdf)
  3. Marzieh Razavi and Mathew Magimai.-Doss. Posterior-Based Multi-Stream Formulation To Combine Multiple Grapheme-to-Phoneme Conversion Techniques. Idiap Research Report, Idiap-RR-33-2015, 2015. (pdf)
  4. Dimitri Palaz, Mathew Magimai.-Doss and Ronan Collobert. Learning linearly separable features for speech recognition using convolutional neural networks. Idiap Research Report, Idiap-RR-24-2015, 2015. (Peer reviewed and presented at ICLR 2015)
  5. Marzieh Razavi, Ramya Rasipuram and Mathew Magimai.-Doss. On the application of automatic subword unit derivation and pronunciation generation for under-resourced language ASR: a study on Scottish Gaelic. Idiap Research Report, Idiap-RR-13-2015, 2015.
  6. Dimitri Palaz, Ronan Collobert and Mathew Magimai.-Doss. End-to-end Phoneme Sequence Recognition using Convolutional Neural Networks. Idiap Research Report, Idiap-RR-40-2013, 2013. (Peer reviewed and presented at NIPS Deep learning Workshop 2013)
  7. Ramya Rasipuram and Mathew Magimai.-Doss. Probabilistic Lexical Modeling and Grapheme-based Automatic Speech Recognition. Idiap Research Report, Idiap-RR-14-2013, 2013.
  8. Ramya Rasipuram and Mathew Magimai.-Doss. KL-HMM and Probabilistic Lexical Modeling. Idiap Research Report, Idiap-RR-04-2013, 2013.
  9. Serena Soldo and Mathew Magimai.-Doss. Integrating Posterior Features and Self-Organizing Maps for Isolated Word Recognition without Dynamic Programming. Idiap Research Report, Idiap-RR-17-2012, 2012.
  10. Anindya Roy, Mathew Magimai.-Doss and Sebastien Marcel. Continuous Speech Recognition using Boosted Binary Features. Idiap Research Report, Idiap-RR-35-2011, 2011.
  11. Ramya Rasipuram and Mathew Magimai.-Doss. Multitask Learning to Improve Articulatory Feature Estimation and Phoneme Recognition. Idiap Research Report Idiap-RR-21-2011, 2011.
  12. Mathew Magimai.-Doss, Fabio Valente, Joel Pinto, Suman Ravuri, and Wen Wang. An investigation on genre-dependent acoustic modeling using MLP features for Mandarin ASR. Idiap Internal Report, Idiap-Internal-RR-179-2010, 2010.
  13. Serena Soldo, Mathew Magimai.-Doss, Joel Pinto, and Hervé Bourlard. On MLP-based posterior features for template-based ASR. Idiap Research Report Idiap-RR-37-2009, 2009.
  14. Mathew Magimai.-Doss, Guillermo Aradilla, and Hervé Bourlard. On joint modelling of grapheme and phoneme information using KL-HMM for ASR. Idiap Research Report Idiap-RR-24-2009, 2009.
  15. Marianna Pronobis and Mathew Magimai.-Doss. Analysis of F0 and cepstral features for robust automatic gender recognition. Idiap Research Report Idiap-RR-30-2009, 2009.
  16. Marianna Pronobis and Mathew Magimai.-Doss. Integrating audio and vision for robust automatic gender recognition. Idiap Research Report Idiap-RR-73-2008, 2008.
  17. Weifeng Li, Mathew Magimai.-Doss, and John Dines. Robust overlapping speech recognition based on neural networks. Idiap Research Report Idiap-RR-55-2007, 2007.
  18. Guillaume Lathoud, Mathew Magimai.-Doss, and Hervé Bourlard. Unsupervised spectral subtraction for noise-robust ASR on unknown transmission channels. Idiap Research Report Idiap-RR-09-2006, 2006.
  19. Mathew Magimai.-Doss, John Dines, Hervé Bourlard, and Hynek Hermansky. Improving continuous speech recognition system performance with grapheme modelling. Idiap Research Report Idiap-RR-16-2005, 2005.
  20. Mathew Magimai.-Doss, John Dines, Hervé Bourlard, and Hynek Hermansky. Phoneme vs grapheme based automatic speech recognition. Idiap Research Report Idiap-RR-48-2004, 2004.
  21. Mathew Magimai-Doss, Todd A. Stephenson, and Hervé Bourlard. Modelling auxiliary information (pitch frequency) in hybrid HMM/ANN based ASR systems. Idiap Research Report Idiap-RR-62-2002, 2002.
  22. Mathew Magimai-Doss and Hervé Bourlard. Pronunciation models and their evaluation using confidence measures. Idiap Research Report Idiap-RR-29-2001, 2001.
  23. Todd A. Stephenson, Mathew Magimai-Doss, and Hervé Bourlard. Automatic speech recognition using pitch information in dynamic Bayesian networks. Idiap Research Report Idiap-RR-41-2000, 2000.