Title :
Minimum Phone Error model training on merged acoustic units for transcribing bilingual code-switched speech
Author :
Ching-Feng Yeh ; Yiu-Chang Lin ; Lin-Shan Lee
Author_Institution :
Grad. Inst. of Commun. Eng., Nat. Taiwan Univ., Taipei, Taiwan
Abstract :
This paper proposes to perform Minimum Phone Error (MPE) model training on merged acoustic units for transcribing Mandarin-English code-switched lectures with highly imbalanced language distribution. Some of the acoustic events in Mandarin and English may have very similar characteristics, so the states or Gaussian mixtures representing them can be merged with identical shared parameters. When MPE is performed afterwards, these merged identical states or Gaussian mixtures can form a compact acoustic unit set. In this way MPE can better discriminate the acoustic units of both languages, because similar units are merged while distinct units are differentiated. Significant improvements in recognition accuracy were observed in the preliminary experiments on real-world bilingual code-switched lecture corpus recorded at National Taiwan University.
Keywords :
Gaussian processes; speech coding; Gaussian mixtures; MPE model training; Mandarin-English code-switched lectures; National Taiwan University; acoustic events; bilingual code-switched speech; compact acoustic unit set; high imbalanced language distribution; minimum phone error model training; real-world bilingual code-switched lecture corpus; recognition accuracy; Accuracy; Acoustics; Hidden Markov models; Merging; Speech; Speech recognition; Training; MPE; bilingual; code-switching; discrimina-tive; merging;
Conference_Titel :
Chinese Spoken Language Processing (ISCSLP), 2012 8th International Symposium on
Conference_Location :
Kowloon
Print_ISBN :
978-1-4673-2506-6
Electronic_ISBN :
978-1-4673-2505-9
DOI :
10.1109/ISCSLP.2012.6423531