Whispered speech recognition based on gammatone filterbank cepstral coefficients
- Авторлар: Marković B.1, Galić J.1, Grozdić Ð.1, Jovičić S.1, Mijić M.1
-
Мекемелер:
- Telecommunication Department, School of Electrical Engineering
- Шығарылым: Том 62, № 11 (2017)
- Беттер: 1255-1261
- Бөлім: Theory and Methods of Signal Processing
- URL: https://journals.rcsi.science/1064-2269/article/view/198953
- DOI: https://doi.org/10.1134/S1064226917110134
- ID: 198953
Дәйексөз келтіру
Аннотация
This paper presents the results on whispered speech recognition using gammatone filterbank cepstral coefficients for speaker dependent mode. The isolated words used for this experiment are taken from the Whi-Spe database. Whispered speech recognition is based on dynamic time warping and hidden Markov models methods. The experiments are focused on the following modes: normal speech, whispered speech and their combinations (normal/whispered and whispered/normal). The results demonstrated an important improvement in recognition after application of cepstral mean subtraction, especially in mixed train/test scenarios.
Авторлар туралы
B. Marković
Telecommunication Department, School of Electrical Engineering
Хат алмасуға жауапты Автор.
Email: brankomarko@yahoo.com
Сербия, Belgrade, 11000
J. Galić
Telecommunication Department, School of Electrical Engineering
Email: brankomarko@yahoo.com
Сербия, Belgrade, 11000
Ð. Grozdić
Telecommunication Department, School of Electrical Engineering
Email: brankomarko@yahoo.com
Сербия, Belgrade, 11000
S. Jovičić
Telecommunication Department, School of Electrical Engineering
Email: brankomarko@yahoo.com
Сербия, Belgrade, 11000
M. Mijić
Telecommunication Department, School of Electrical Engineering
Email: brankomarko@yahoo.com
Сербия, Belgrade, 11000