Automatic Segmentation of Indonesian Speech into Syllables using Fuzzy Smoothed Energy Contour with Local Normalization, Splitting, and Assimilation

Abstract

Abstract. This paper discusses the usage of the short-term energy contour of speech smoothed by a fuzzy-based method to automatically segment it into syllabic units. Two new additional procedures, local normalization and postprocessing, are proposed to adapt to the Indonesian language. Testing to 220 Indonesian utterances showed that the local normalization significantly improved the performance of the fuzzy-based smoothing. In the postprocessing procedure, splitting and assimilation work in different ways. The splitting of missed short syllables sharply reduced deletion, but slightly increased insertion. On the other hand, the assimilation of a single consonant segment into an expected previous or next segment slightly reduced insertion, but increased deletion. The use of splitting gave a higher accuracy than the assimilation and combined splittingassimilation procedures, since in many cases the assimilation keeps the unexpected insertions and overmerges the expected segments.Keywords: assimilation, fuzzy-based smoothing; Indonesian language; local normalization; short-term energy contour; splitting; syllable segmentation. IntroductionInformation on syllabic units can be used to improve the performance of flat start-based automatic speech recognition (ASR) [1]- [11]. In 2010, Janakiraman et al. [11] reported that incorporating information on syllable boundaries into English ASR reduced both computational complexity and word error rate (WER) significantly compared to flat start ASR. The WER can be reduced from 13% to 4.4% and from 36% to 21.2% for TIMIT and NTIMIT databases respectively.Every language has unique characteristics. For example, English and Indonesian have different syllable patterns. A study of telephone conversations and switchboard corpus by has shown that English has 80% monosyllabic words and 85% of them are simple structures (V, VC, CV, CVC)

Optimizing Parameters of Automatic Speech Segmentation into Syllable Units

“…In general, there are three metrics used to evaluate a speech segmentation, i.e. accuracy, insertion error, and deletion error [3]. The accuracy is the ratio between number of correct segment and the total segment in the sentence.…”

“…In [3], an Indonesian syllable segmentation is developed using an STE smoothed by a fuzzy-based model. A local normalization and a fixed threshold are well designed to improve the system.…”

“…There are two approaches of speech segmentation, i.e. time-domain approach [3,4] and frequency-domain approach [5][6][7]. The time-domain approach, which is commonly based on a Short-Term Energy (STE), requires a low computation but less accuracy than the frequencydomain one.…”

“…However, the STE-based approach is good enough for defining syllable boundaries since a high energy in STE generally corresponds to a vowel utterance that is a nucleus of syllable. Hence, many researchers develop a syllable segmentation based on an STE [3,4].…”

“…The problem of an STE-based syllable segmentation is that it needs a fixed threshold that should be determined using a manual observation as described in [3,4]. In this paper, an adaptive threshold method that is automatically inducted by learning a set of utterances is proposed to detect a syllable boundary…”