Basis Identification for Automatic Creation of Pronunciation Lexicon for Proper Names

TL;DR

This paper introduces an optimization-based method to automatically generate a pronunciation lexicon for proper names by constructing a small, orthogonal basis set of words that can span the entire name database, reducing manual effort.

Contribution

It proposes two novel algorithms for basis construction that enable automatic pronunciation lexicon creation, validated on large name datasets with language-independent approaches.

Findings

Transcription achieved by transcribing a small basis set

Algorithms are effective across different languages

Performance improves with names of similar origin

Abstract

Development of a proper names pronunciation lexicon is usually a manual effort which can not be avoided. Grapheme to phoneme (G2P) conversion modules, in literature, are usually rule based and work best for non-proper names in a particular language. Proper names are foreign to a G2P module. We follow an optimization approach to enable automatic construction of proper names pronunciation lexicon. The idea is to construct a small orthogonal set of words (basis) which can span the set of names in a given database. We propose two algorithms for the construction of this basis. The transcription lexicon of all the proper names in a database can be produced by the manual transcription of only the small set of basis words. We first construct a cost function and show that the minimization of the cost function results in a basis. We derive conditions for convergence of this cost function and validate them experimentally on a very large proper name database. Experiments show the transcription can be achieved by transcribing a set of small number of basis words. The algorithms proposed are generic and independent of language; however performance is better if the proper names have same origin, namely, same language or geographical region.