Exploiting speech source information for vowel landmark detection for low resource language

Landmarks are the time instants in a speech signal which marks important events (such as vowels, consonants and glides) in the speech signal. This paper proposes use of novel vowel landmark detection (VLD) algorithm for low resourced language, viz., Gujarati, an Indian language. The proposed VLD method uses speech source information to detect the vowel landmarks which are points of high sonority. The excitation peaks in Hilbert envelope of Teager energy profile of zero frequency filtered (ZFF) speech signal can be interpreted as perceptually significant feature which contribute to the loudness. The performance of proposed VLD method is compared with existing loudness-based method. The results are reported on speech recorded in three different modes, viz., read, spontaneous and lecture followed by manual phonetic transcription by the transcribers (to be used as the ground truth) for Gujarati. In particular, the proposed VLD algorithm performs relatively better than an existing loudness-based method. The proposed VLD algorithm has detection rate of 78.92 %, 76.40 % and 73.89 %, which is 8.79 %, 7.23 % and 7.17 % more as compared to loudness-based method in lecture, spontaneous and read mode, respectively. The proposed algorithm is also shown to be robust against signal degradations such as white noise. In addition, proposed algorithm is fast and requires no training.