Probabilistic Approach to Arabic Speech Correction for Peoples with Language Disabilities

2015 ◽  
Vol 5 (4) ◽  
pp. 1-18 ◽  
Author(s):  
Naim Terbeh ◽  
Mohamed Achraf Ben Mohamed ◽  
Mounir Zrigui
Keyword(s):  
Probabilistic Approach ◽  
Acoustic Model ◽  
Probabilistic Representation ◽  
Large Vocabulary ◽  
Probability Of Occurrence ◽  
Correction Rate ◽  
Language Disabilities ◽  
Speech Correction ◽  
Standard Probability ◽  
Correction System

This work consists in achieving an automatic speech correction system for continuous Arabic speech with large vocabulary in mono-speaker mode. Two vectors to be generated: the first is an Arabic speech standard (probability of occurrence of each Arabic bi-phoneme), the second gives a probabilistic representation of the speech sequence to be corrected. Using these two vectors, phonemes that pose pronunciation problems to speakers and their replacements can be identified. The rest is a game of substitutions and belonging tests to an Arabic lexicon. For that, an acoustic model for Arabic speech and a lexicon of 4 million distinct words have been built. Results of the work were encouraging and present a reference for other works for people with language disabilities. A correction rate of 97% is reached.

scholarly journals Probabilistic coastal vulnerability assessment to storms at regional scale – application to Catalan beaches (NW Mediterranean)

2011 ◽  
Vol 11 (2) ◽  
pp. 475-484 ◽  
Author(s):  
E. Bosom ◽  
J. A. Jiménez
Keyword(s):  
Probability Distributions ◽  
Regional Scale ◽  
Probabilistic Approach ◽  
Wave Climate ◽  
Coastal Vulnerability ◽  
Large Variability ◽  
Probability Of Occurrence ◽  
Nw Mediterranean ◽  
Offshore Wave ◽  
Extreme Probability

Abstract. A methodology to assess storm-induced coastal vulnerability taking into account the different induced processes separately (inundation and erosion) is presented. It is based on a probabilistic approach where hazards time series are built from existing storm data and later used to fit an extreme probability function. This is done for different sectors along the coast defined in terms of the wave climate and for representative beach types of the area to be analyzed. Once probability distributions are available, coastal managers must decide the probability of occurrence to be accepted as well as the period of concern of the analysis in function of the importance of the hinterland. These two variables will determine the return period to be considered in the assessment. The comparison of hazards and vulnerabilities associated with the selected probability of occurrence permit to identify the most hazardous areas along the coast in a robust manner by including the spatial variability in forcing (storm climate) and receptor (beaches). The methodology has been applied to a 50 km long coastal stretch of the Catalonia (NW Mediterranean) where offshore wave conditions can be assumed to be homogeneous. In spite of this spatially constant wave field, obtained results indicate a large variability in hazards intensity and vulnerability along the coast.

scholarly journals Acoustic model topology optimization for large vocabulary speech recognition

2020 ◽  
Vol 309 ◽  
pp. 05007
Author(s):  
Xirimo Bao ◽  
Chunmei Ning
Keyword(s):  
Speech Recognition ◽  
Recognition Performance ◽  
Selection Method ◽  
Acoustic Model ◽  
Swarm Optimization ◽  
Large Vocabulary ◽  
Recognition Systems ◽  
Large Vocabulary Speech Recognition ◽  
Model Topology ◽  
Topology Selection

Acoustic model topology selection work in constructing large vocabulary speech recognition systems is being done empirically or heuristically. In this paper, we propose two improved algorithms, which are based on Genetic Algorithm (GA) and Particle Swarm Optimization (PSO) respectively, on the basis of our previously proposed algorithms to select and optimize model topologies for small or medium vocabulary speech recognition systems. Our improved algorithms attain the goal of optimizing acoustic model topologies for large vocabulary speech recognition systems mainly through modifying the encoding schemes of our previously proposed algorithms. Experiments on the dialogue corpus of Inner Mongolia University show that, compared with the conventional acoustic model topology selection method, our newly proposed algorithms are able to bring much higher recognition performance for large vocabulary speech recognition systems by optimizing their acoustic model topologies.

Sign in / Sign up

Export Citation Format

Share Document