Segmentation of speech into syllable units using fuzzy smoothed short term energy contour

Sheikhi, Ghazaal; Almasganj, Farshad

doi:10.1109/icbme.2011.6168554

Cited by 7 publications

(14 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Sheikhi and Almasganj have shown that the Teager energy gives the best accuracy for the Farsi speech dataset [12]. However, in this research, the square energy as in Eq.…”

Section: Short-term Energymentioning

confidence: 82%

“…Hence, local normalization is applied to the STE contour by detecting frames of very low energy and then normalizing the set of high energy frames that occur between those very low energy frames to the maximum energy in that set. This step is expected to produce a better STE contour than that produced by the global normalization used in [12].…”

Section: Local Normalizationmentioning

confidence: 99%

“…1) the time domain, as in [12], [15]- [17]; 2) the frequency domain, as in [11], [18]- [25]; and a combination of both, as in [26]- [28]. The time domain approaches mostly use the short-term energy (STE) contour smoothed by a smoothing algorithm, while the frequency domain approaches exploit cepstrum features.…”

Section: Related Workmentioning

confidence: 99%

“…This method works very well but only for short-sentence utterances. In [12], the plain STE contour was first smoothed by fuzzy-based smoothing before defining the syllable boundaries using a threshold-based method. The usage of fuzzy-based smoothing gave a much higher accuracy, 93.8% for a Farsi speech dataset, than the common moving average smoothing method.…”

Section: Related Workmentioning

confidence: 99%

“…A segmentation method in [12] that was designed for the Farsi language, with simple syllable structures CV(C)(C), was adapted and tested to the Indonesian speech dataset of the clean speech corpus as described in [13]. Some modifications and two additional procedures, i.e.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

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

Suyanto¹,

Putra²

2014

J. ICT.Res.Appl.

View full text Add to dashboard Cite

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)

show abstract

“…Sheikhi and Almasganj have shown that the Teager energy gives the best accuracy for the Farsi speech dataset [12]. However, in this research, the square energy as in Eq.…”

Section: Short-term Energymentioning

confidence: 82%

Section: Local Normalizationmentioning

confidence: 99%