Boundary detection using Continuous Wavelet Analysis
نویسندگان
چکیده
Unsupervised boundary detection and classification is both a theoretically interesting question and an important challenge for speech technology. Theoretical interest lies in exploring how and to what extent is the boundary information encoded in purely acoustic material. For technology, automatic boundary detection facilitates cheap and fast labeling of large corpora of speech data. In this work we present a novel methodology of automatic and unsupervised boundary detection and classification based on the continuous wavelet transform (CWT) technique. Several approaches using lines of minimal amplitude, phase information and wavelet-based estimation of speech tempo are evaluated and compared on Boston Radio News Corpus data. The results show that this methodology using hierarchical information encoded in speech signal compares favorably with traditionally used supervised boundary detection techniques using acoustic information.
منابع مشابه
Subsea Free Span Pipeline Damage Detection Based on Wavelet Transform under Environmental Load
During their service life, marine pipelines continually accumulate damage as a result of the action of various environmental forces. Clearly, the development of robust techniques for early damage detection is very important to avoid the possible occurrence of a disastrous structural failure. Most of vibration-based damage detection methods require the modal properties that are obtained from mea...
متن کاملFault Strike Detection Using Satellite Gravity Data Decomposition by Discrete Wavelets: A Case Study from Iran
Estimating the gravity anomaly causative bodies boundary can facilitate the gravity field interpretation. In this paper, 2D discrete wavelet transform (DWT) is employed as a method to delineate the boundary of the gravity anomaly sources. Hence, the GRACE’ satellite gravity data is decomposed using DWT. DWT decomposites a single approximation coefficients into four distinct components: the appr...
متن کاملDamage identification of structures using experimental modal analysis and continuous wavelet transform
Abstract: Modal analysis is a powerful technique for understanding the behavior and performance of structures. Modal analysis can be conducted via artificial excitation, e.g. shaker or instrument hammer excitation. Input force and output responses are measured. That is normally referred to as experimental modal analysis (EMA). EMA consists of three steps: data acquisition, system identificatio...
متن کاملReservoir Rock Characterization Using Wavelet Transform and Fractal Dimension
The aim of this study is to characterize and find the location of geological boundaries in different wells across a reservoir. Automatic detection of the geological boundaries can facilitate the matching of the stratigraphic layers in a reservoir and finally can lead to a correct reservoir rock characterization. Nowadays, the well-to-well correlation with the aim of finding the geological l...
متن کاملApplication of Wavelet Transform as a Signal Processing Method for Defect Detection using Lamb Waves: Experimental Verification
A Lamb wave-based crack detection method for aluminum plates health monitoring is developed in this paper. Piezoelectric disks are employed to actuate and capture the Lamb wave signals. The position of crack is assumed to be aligned with the sensor and actuator. Extraction of high quality experimental results of lamb wave propagation in a plate-like structure is considerably complicated due to...
متن کامل