Since rule based systems for specific domains require their knowledge sources to be manually revised when ported to other domains, we enrich our rule based recognizer and turn it into a hybrid recognizer so that it learns from annotated data when available and improves its knowledge sources accordingly.
To do so, a Levenshtein distance-based word identifier computes the Levenshtein distance between the phone sequence given by the phone-based recognizer and the phone transcription of the OOV term so that a low Levenshtein distance match suggests the occurrence of this OOV term.
In this study, we present a hybrid named entity recognizer for Turkish, which is based on a manually engineered rule based recognizer that we have proposed.
In "Novel kernel based recognizers of human actions," Danafar et al. study unsupervised and supervised recognition of human actions in video sequences.
This GMM-HMM word matcher takes the phone transcription given by the phone based speech recognizer in the time intervals corresponding to the start and end times of the segment, and produces the most likely word/s given this phone transcription.
To handle these problems, using the GMM based HMM recognizer for discriminating the non-speech from the speech not only can reduce the number of mixtures but also can improve the accuracy of VAD without the experimental threshold.
The potential of ART1 based pattern recognizer to recognize patterns, monochrome and colour, noisy and noise free has been studied on two experimental problems.
For the speech recognizer and the semantic analysis modules, we use software developed by IBM and CSLR Center for Spoken Language Researchh at University of Colorado), respectively.
A Mandarin speech recognizer, using hidden Markov model toolkit (HTK) [36, 37], was implemented to construct the keyword recognizer and the possible outcomes were analyzed for miscommunication handling.
The other is based on a phone recognizer and a search in the 1-best phone sequence from an edit distance approach.
The final decision of the face recognizer is based on the weighted sum of the errors computed from each of the regions.
