Using of Beylkin Wavelet for Speech Recognition

This paper describes the application of the Beylkin wavelet for speech segmentation. The problem of speech segmentation in the Yakut language is that there are segmentation difficulties due to the peculiarities of the language. The use of long vowels and double consonants in the Yakut language complicates the correct segmentation of oral speech. For the analysis, the window method of analyzing the energy of the wavelet signal is used. The experience of using different wavelet functions has shown that it is not always possible to accurately find the segment boundaries in some cases. The Scilab package has a large library of wavelets that allows extensive research into their applications in speech recognition. The results of the study show that there are difficulties due to various reasons, one of which is the presence of double sonorant consonants. The graphs of the analysis of doubled sonorant consonants are given.

3 Infant-directed Vocal Communication in Sheep

Infant-directed speech (IDS) in humans, AKA motherese, is different from normal speech with a higher pitch, higher frequency range, slower pace, and more repetition. infants usually are believed to react differently to IDS compared to adult-directed speech. Studies showed that IDS facilitates infant’s speech segmentation, word memory, word learning, and communicative development. IDS is common across languages and cultures, but the evolutionary origin of IDS is a myth. The objective of this study is to find out whether the special style of vocalization namely infant-directed vocalization (IDV), which differs from adult-directed vocalization (ADV), can be also observed in non-human, even non-primate species. The ADV and IDV of ewes were recorded. The sound wave features of the recordings were analyzed by visualization and machine learning. The ADV had representative peak frequencies at 175, 720, and 860Hz, while IDV only had one peak characteristic frequency at 245Hz. The machine-learning algorithm was able to clearly identify (overall accuracy was 89.3%) the distinguishing characteristics between ADV and IDV. Then we tested if the lamb reacts differently to the ewe’s IDV and ADV. The recording was played when the pre-weaning lambs were individually kept and the behavior of the lambs was recorded. The results showed that the lambs looked towards the sound source when IDV was played more than ADV (6.1 vs 3.1 times/5 min); they moved towards the sound source of IDV 8.6 times per 5 min compared to ADV which was 2.8 times/5min), and they bleated back to the sound source when IDV was played (18.0 times/5 min) more than when ADV was played (11.3 time/5 min); within 2 min after the recording played, lambs bleated back to IDV 8 times compared to ADV 4.8 times. This indicated the ewes’ IDV and ADV show different socio-emotional and attention effects on their lambs.