DocumentCode
390501
Title
Spectral analysis of some typical Sino-Tibetan languages with special reference to North East Indian languages
Author
Talukdar, P.H. ; Sarma, Shikhar Kr
Author_Institution
Univ. Sci. Instrum. Centre, Gauhati Univ., Assam, India
Volume
1
fYear
2002
fDate
26-30 Aug. 2002
Firstpage
520
Abstract
We have made an attempt to study the spectral characteristics of two North East Indian languages, Assamese and Boro, coming from different genres. We have taken a few words with similar, partially similar, and dissimilar characteristics in their nature of utterance from Assamese and Boro. The spectral analysis revels that both the languages have a high degree of similarity at both voiced and unvoiced regions for words with the same meaning. LPC modeling of the random selection of similar, partially similar and dissimilar phonemes may not provide characteristic features of the two languages, but cepstral coefficient analysis of the LPC modeling may provide important information in studying the linguistic and phonetic distinction of the two languages and, finally, speaker identification and word verification.
Keywords
cepstral analysis; linear predictive coding; linguistics; natural languages; speech processing; Assamese; Boro; LPC modeling; North East Indian languages; Sino-Tibetan languages; cepstral coefficient analysis; linguistics; phonetics; speaker identification; spectral analysis; word verification; Artificial neural networks; Cepstral analysis; Hidden Markov models; Instruments; Linear predictive coding; Natural languages; Spectral analysis; Speech analysis; Speech recognition; Training data;
fLanguage
English
Publisher
ieee
Conference_Titel
Signal Processing, 2002 6th International Conference on
Print_ISBN
0-7803-7488-6
Type
conf
DOI
10.1109/ICOSP.2002.1181107
Filename
1181107
Link To Document