Several nontrivial applications of subword trees have been developed since their first appearance. Some stich applications depart considerably from the original motivations. A brief account of them is attempted here. INTRODUcnON Subword trees fit in the general subject of digitaEearch indexes [KNj. In fact their earliest conception is somewhat implicit in Morrison"s 'PATRICIA' tries [MOl. Sever...

In this paper, an unexplored subject in the domains of Farsi/Arabic handwritten word preprocessing is introduced. Subwords play a vital role in many applications such as cheque amount recognition, text recognition, lexicon reduction and subword-based word recognition. Correcting the faults occurred in subwords will improve the overall performance of these applications. A subword is a connected-...

For languages with limited training resources, out-ofvocabulary (OOV) words are a significant problem, both for transcription and keyword spotting. This paper investigates the use of subword lexical units for keyword spotting. Three strategies for using the sub-word units are explored: 1) converting word-based lattices to subword lattices after decoding, 2) performing a separate decoding for ea...

In this work, we describe a subword unit approach for information retrieval of items by voice. An algorithm based on the minimum description length (MDL) principle converts an index written in terms of words into an index written in terms of phonetic subword units. A speech recognition engine that uses a language model and pronounciation dictionary built from such an inventory of subword units ...

In this paper, we use subword modeling to learn the pronunciations and spellings of new words. The subwords are generated with a context-free grammar, and are intermediate units between phonemes and syllables. We first evaluate the effectiveness of the subword model in automatically generating the spelling and pronunciation of new words. Then the subword model is embedded in a multi-stage recog...

Speaker variability has a significant impact to the state-of-the-art speech recognition systems. Traditionally speaker clustering is performed without considering individual or class phonetic similarities across different speakers. In fact, clustered speaker groups may have very different degrees of variations for different phonetic classes. In this paper, speaker clustering is performed at sub...

We describe a canonical spanning tree of the ridge graph of a subword complex on a finite Coxeter group. It is based on properties of greedy facets in subword complexes, defined and studied in this paper. Searching this tree yields an enumeration scheme for the facets of the subword complex. This algorithm extends the greedy flip algorithm for pointed pseudotriangulations of points or convex bo...