CONF Imseng_ICASSP_2011/IDIAP Language dependent universal phoneme posterior estimation for mixed language speech recognition Imseng, David Bourlard, Hervé Magimai-Doss, Mathew Dines, John https://publications.idiap.ch/index.php/publications/showcite/Imseng_Idiap-RR-13-2011 Related documents Proceedings IEEE International Conference on Acoustics, Speech and Signal Processing Prag, CZ 2011 5012-5015 REPORT Imseng_Idiap-RR-13-2011/IDIAP Language dependent universal phoneme posterior estimation for mixed language speech recognition Imseng, David Bourlard, Hervé Magimai-Doss, Mathew Dines, John Idiap-RR-13-2011 2011 Idiap May 2011 This paper presents a new approach to estimate "universal" phoneme posterior probabilities for mixed language speech recognition. More specifically, we propose a new theoretical framework to combine phoneme class posterior probabilities in a principled way by using (statistical) evidence about the language identity. We investigate the proposed approach in a mixed language environment (SpeechDat(II)) consisting of five European languages. Our studies show that the proposed approach can yield significant improvements on a mixed language task, while maintaining the performance on monolingual tasks. Additionally, through a case study, we also demonstrate the potential benefits of the proposed approach for non-native speech recognition.