Automatic Speech Segmentation Using Hybrid Wavelet Features and HMM

Amr Gody,Amr Saleh,Manal Shabaan

doi:10.21608/ejle.2016.60172

Amr Gody, Amr Saleh + Show 1 more

Open Access

https://doi.org/10.21608/ejle.2016.60172

Copy DOI

Abstract

In this research, a novel feature set is used to automatically segment speech signal. Automatic segmentation is veryuseful especially for large database. A hybrid features model is created from wavelet packet analysis and mel-scale is used to train Hidden Markov Model (HMM) for phone boundary detection. HMM is implemented using the Hidden Markov Model Toolkit (HTK).The database (Ked-TIMIT) is used for result verifications and Mel Frequency Cepstral Coefficients (MFCC) is used as reference for evaluating the results of the proposed Hybrid model. The results are categorized for vowels, consonants and short phones. Phone duration and start location are used as metrics to evaluate the system success rate. Success rate of 74% is achieved for consonant detection, 72% for vowel detection and 58% for short phone detection. Using the simple metric that relies only on boundary locations but ignoring duration, the achieved results are 92.5% for consonant detection, 90% for vowel detection and 77.5% for short phoneme detection. In addition to boundary detection the proposed hybrid model is utilized to compare newly developed features called Mel scale Best Tree Encoding (Mel-BTE ) to the mostly used popular features MFCC along with all experiments using the same database. The relative results for Mel-BTE with respect to MFCC are 94.77% for consonant detection, 87.5% for vowel detection and 93.33% for short phoneme detection.

Highlights

Phone segmentation is a process of finding the boundaries of a sequence of known phones in a spoken utterance
The proposed method achieves 86.8% phoneme boundary detection accuracy at framelevel when tested on TIMIT database
It is indicated that through this research paper the automatic segmentation problem can be altered using some hybrid techniques that are related to spectrum analysis and statistical model

Summary

Introduction

Phone segmentation is a process of finding the boundaries of a sequence of known phones in a spoken utterance. Phone segmentation process is still an active topic, as is shown by the range of research directions suggested . Is part of the table from [2] that includes the best results and the best technique according to their experimental results. The obtained results using the proposed hybrid model is about 91%, which is competitive to the achieved results in [2] but in this research paper the model is much simpler than the proposed model in [2]

Methods

Results

Conclusion