Abstract: Speech segmentation techniques have made important advances in the past decades and are still an active area of research and development. It is a process of breaking down a speech signal into smaller units such as phonemes. Speech segmentation is decisive for many acoustic systems essentially Automatic Speech Recognition (ASR). Phonetic segmentation techniques are divided into two major categories: Text-Dependent (TD) and Text-Independent (TI). In the text-dependent segmentation techniques, the phonetic annotation of the speech signal is already known and we only need to find the boundaries of each phoneme segment. While in the Text-Independent (TI) techniques no annotation is available, thus, the segmentation relies solely on the acoustic information contained in the speech signal. In this study, we present a thorough survey of the different algorithms and techniques proposed so far for solving the problem of text-independent phonetic segmentation.
Ihsan Al-Hassani, Oumayma Al-Dakkak and Abdlnaser Assami, 2021. A Survey: Text Independent Automatic Speech Segmentation Techniques. Research Journal of Applied Sciences, 16: 75-84.