Study of Phonemes Confusions in Hierarchical Automatic Phoneme Recognition System
This work addresses robustness issues in phoneme recognition for speech processing applications, but it appears incremental as it builds on existing confusion analysis methods.
The paper tackled the problem of phoneme confusions affecting recognition rates in automatic phoneme recognition systems by analyzing confusion matrices and pronunciation similarities, resulting in a new hierarchical recognizer that showed significant improvements on the TIMIT database.
In this paper, we have analyzed the impact of confusions on the robustness of phoneme recognitions system. The confusions are detected at the pronunciation and the confusions matrices of the phoneme recognizer. The confusions show that some similarities between phonemes at the pronunciation affect significantly the recognition rates. This paper proposes to understand those confusions in order to improve the performance of the phoneme recognition system by isolating the problematic phonemes. Confusion analysis leads to build a new hierarchical recognizer using new phoneme distribution and the information from the confusion matrices. This new hierarchical phoneme recognition system shows significant improvements of the recognition rates on TIMIT database.