In this paper, we propose a novel approach to estimate three types of phone mismatch penalty matrices for two-state keyword spotting. When the output of a phone recognizer is given, text matching with the phone sequences provided by the specified keyword using the proposed phone mismatch penalty matrices is carried out to detect a specific keyword. The penalty matrices which is estimated from t...

It is one of most essential issues to extract the keywords from conversational speech for understanding the utterances from speakers. This thesis aims at keyword spotting from spontaneous speech for keyword detecting. We proposed prosodic features that are used for keyword detection. The prosody words are segmented from speaker’s utterance according to the pre-training decision tree. The suppor...

This paper proposes a novel method of generating statistical Korean Hangul character models in real time. From a set of grapheme average images we compose any character images, and then convert them to P2DHMMs. The nonlinear, 2D composition of letter models in Hangul is not straightforward and has not been tried for machine-print character recognition. It is obvious that the proposed method of ...

In this paper we present a word spotting system in text lines for offline Indic scripts such as Bangla (Bengali) and Devanagari. Recently, it was shown that zone-wise recognition method improves the word recognition performance than conventional full word recognition system in Indic scripts [29]. Inspired with this idea we consider the zone segmentation approach and use middle zone information ...

The ever-increasing volume of audio data available online through the world wide web means that automatic methods for indexing and search are becoming essential. Hidden Markov model (HMM) keyword spotting and lattice search techniques are the two most common approaches used by such systems. In keyword spotting, models or templates are defined for each search term prior to accessing the speech a...

Normally, we represent speech as a long sequence of frames and model the keyword with a relatively small set of parameters, commonly with a hidden Markov model (HMM). However, since the input speech is much longer than the keyword, suppose instead that we represent the speech as a relatively sparse set of impulses (roughly one per phoneme) and model the keyword as a filter-bank where each filte...

The papers in this session focus on techniques for and applications of large-vocabulary continuous speech recognition. The technique oriented papers discuss techniques for channel compensation, fast search, acoustic modeling, and adaptive language modeling. The applications oriented papers discuss methods for using recognizers for language identification, speaker identification, speakersex iden...

Word spotting is the process of retrieving all instances of a queried keyword from a digital library of document images. In this paper we evaluate the performance of different word descriptors to assess the advantages and disadvantages of statistical and structural models in a framework of query-by-example word spotting in historical documents. We compare four word representation models, namely...

This paper explores one dimension along which word spotting and speech recognition differ: the nature of the background model. In word spotting, a relatively small number of keywords float on a sea of unknown words. In speech recognition, an occasional unknown word punctuates utterances that are otherwise completely invocabulary. Despite this difference in viewpoint, in some circumstances imple...