5th International Conference on Spoken Language Processing
High OOV-rates are one of the most prevailing problems for languages with a rapid vocabulary growth, e.g. when transcribing Serbo-Croatian and German broadcast news. Hypothesis-Driven-Lexical-Adaptation (HDLA) has been shown to decrease high OOV-rates significantly by using morphology-based linguistic knowledge. This paper introduces another approach to dynamically adapt a recognition lexicon to the utterance to be recognized. Instead of morphological knowledge about word stems and inflection endings, distance measures based on Levenstein distance are used. Results based on phoneme and grapheme distances will be presented. Our distance-based approach requires no expert knowledge about a specific language and no definition of complex grammar rules. Instead, grapheme sequences or the phoneme representation of words are sufficient to apply our HDLA-algorithm easily to any new language. With our proposed technique OOV-rates were decreased by more than half from 8.7% to 4%, thereby also improving recognition performance by an absolute 4.1% from 29.5% to 25.4% word error rate.
Bibliographic reference. Geutner, Petra / Finke, Michael / Waibel, Alex (1998): "Phonetic-distance-based hypothesis driven lexical adaptation for transcribing multlingual broadcast news", In ICSLP-1998, paper 0771.