This paper presents methods and results for optimizing subword detectors in continuous speech. Speech detectors are useful within areas like detection-based ASR, pronunciation training, phonetic analysis, word spotting, etc. We build detectors for both articulatory features and phones by discriminative training of detector-specific MFCC filterbanks and HMMs. The resulting filterbanks are clearly different from each other and reflect acoustic properties of the corresponding detection classes. For the TIMIT task, our detector-specific features reduce the average detection error rate by 20% compared to standard MFCCs.
Bibliographic reference. Canterla, Alfonso M. / Johnsen, Magne H. (2011): "Optimized feature extraction and HMMs in subword detectors", In INTERSPEECH-2011, 2397-2400.