Structural Analysis of Hindi Phonetics and A Method for Extraction of Phonetically Rich Sentences from a Very Large Hindi Text Corpus

TL;DR

This paper analyzes Hindi phonetics and proposes a two-stage algorithm to extract a minimal set of phonetically rich sentences from a large corpus, improving speech recognition and synthesis training.

Contribution

It introduces a statistical analysis of Hindi phonetics and a novel two-stage selection algorithm for phonetically rich sentences with high triphone diversity.

Findings

Efficiently extracts a minimal, diverse set of sentences covering all phonetic units.

Improves triphone distribution uniformity in the selected sentence set.

Reduces the number of sentences needed for effective speech module training.

Abstract

Automatic speech recognition (ASR) and Text to speech (TTS) are two prominent area of research in human computer interaction nowadays. A set of phonetically rich sentences is in a matter of importance in order to develop these two interactive modules of HCI. Essentially, the set of phonetically rich sentences has to cover all possible phone units distributed uniformly. Selecting such a set from a big corpus with maintaining phonetic characteristic based similarity is still a challenging problem. The major objective of this paper is to devise a criteria in order to select a set of sentences encompassing all phonetic aspects of a corpus with size as minimum as possible. First, this paper presents a statistical analysis of Hindi phonetics by observing the structural characteristics. Further a two stage algorithm is proposed to extract phonetically rich sentences with a high variety of triphones from the EMILLE Hindi corpus. The algorithm consists of a distance measuring criteria to select a sentence in order to improve the triphone distribution. Moreover, a special preprocessing method is proposed to score each triphone in terms of inverse probability in order to fasten the algorithm. The results show that the approach efficiently build uniformly distributed phonetically-rich corpus with optimum number of sentences.