Grapheme-to-phoneme conversion based on high-order Markov chain for spoken term detection by text query
Dmitriy Prozorov, Alexandra Tatarinova · 2017
Spoken term detection (STD) is a fundamental part of some speech processing applications. One of STD methods uses a phoneme representation of words from a spoken content and a text query. The paper presents a new grapheme-to-phoneme conversion method based on high-order Markov chain. The method is applied to retrieve of spoken documents in Russian language. The aim of this research is evaluation effectiveness of STD method based on the proposed grapheme-to-phoneme conversion.