Handbook of the International Phonetic Association: A Guide to the Use of the International Phonetic Alphabet

Different levels of an intelligent video surveillance system (IVVS) are studied in this review.Existing approaches for abnormal behavior recognition relative to each level of an IVVS are extensively reviewed.Challenging datasets for IVVS evaluation are presented.Limitations of the abnormal behavior recognition area are discussed. With the increasing number of surveillance cameras in both indoor and outdoor locations, there is a grown demand for an intelligent system that detects abnormal events. Although human action recognition is a highly reached topic in computer vision, abnormal behavior detection is lately attracting more research attention. Indeed, several systems are proposed in order to ensure human safety. In this paper, we are interested in the study of the two main steps composing a video surveillance system which are the behavior representation and the behavior modeling. Techniques related to feature extraction and description for behavior representation are reviewed. Classification methods and frameworks for behavior modeling are also provided. Moreover, available datasets and metrics for performance evaluation are presented. Finally, examples of existing video surveillance systems used in real world are described.

Abnormal behavior recognition for intelligent video surveillance systems

Language is universal, but it has few indisputably universal characteristics, with cross-linguistic variation being the norm. For example, languages differ greatly in the number of syllables they allow, resulting in large variation in the Shannon information per syllable. Nevertheless, all natural languages allow their speakers to efficiently encode and transmit information. We show here, using quantitative methods on a large cross-linguistic corpus of 17 languages, that the coupling between language-level (information per syllable) and speaker-level (speech rate) properties results in languages encoding similar information rates (~39 bits/s) despite wide differences in each property individually: Languages are more similar in information rates than in Shannon information or speech rate. These findings highlight the intimate feedback loops between languages' structural properties and their speakers' neurocognition and biology under communicative pressures. Thus, language is the product of a multiscale communicative niche construction process at the intersection of biology, environment, and culture.

Different languages, similar encoding efficiency: Comparable information rates across the human communicative niche.

This paper presents our work in automatic speech recognition (ASR) in the context of under-resourced languages with application to Vietnamese. Different techniques for bootstrapping acoustic models are presented. First, we present the use of acoustic-phonetic unit distances and the potential of crosslingual acoustic modeling for under-resourced languages. Experimental results on Vietnamese showed that with only a few hours of target language speech data, crosslingual context independent modeling worked better than crosslingual context dependent modeling. However, it was outperformed by the latter one, when more speech data were available. We concluded, therefore, that in both cases, crosslingual systems are better than monolingual baseline systems. The proposal of grapheme-based acoustic modeling, which avoids building a phonetic dictionary, is also investigated in our work. Finally, since the use of sub-word units (morphemes, syllables, characters, etc.) can reduce the high out-of-vocabulary rate and improve the lack of text resources in statistical language modeling for under-resourced languages, we propose several methods to decompose, normalize and combine word and sub-word lattices generated from different ASR systems. The proposed lattice combination scheme results in a relative syllable error rate reduction of 6.6% over the sentence MAP baseline method for a Vietnamese ASR task.

Automatic Speech Recognition for Under-Resourced Languages: Application to Vietnamese Language

This paper presents our first steps in fast acoustic modeling for a new target language. Both knowledge-based and data-driven methods were used to obtain phone mapping tables between a source language (French) and a target language (Vietnamese). While acoustic models borrowed directly from the source language did not perform very well, we have shown that using a small amount of adaptation data in the target language (one or two hours) lead to very acceptable automatic speech recognition (ASR) performance. Our best continuous Vietnamese recognition system, adapted with only two hours of Vietnamese data, obtains a word accuracy of 63.9% on one hour of Vietnamese speech dialog for instance.

First steps in fast acoustic modeling for a new target language: application to Vietnamese

This paper presents an overview of our activities for spoken and written language resources for Vietnamese implemented at CLIPSIMAG Laboratory and International Research Center MICA. A new methodology for fast text corpora acquisition for minority languages which has been applied to Vietnamese is proposed. The first results of a process of building a large Vietnamese speech database (VNSpeechCorpus) and a phonetic dictionary, which is used for automatic alignment process, are also presented.

/pdf/spoken-and-written-language-resources-for-vietnamese-zom5egk6ly.pdf

Spoken and Written Language Resources for Vietnamese

Understanding and managing tonal characteristics of Vietnamese language is one of the most difficult aspects in Vietnamese speech processing. However, at present, there is no common agreement about the influence of fundamental frequency (F0) on the perception of Vietnamese syllables. Thus, instead of analyzing the F0 of a limited number of Vietnamese syllables like other methods found in literature, this paper will present a new methodology based on, first the synthesis of arbitrary syllables, and secondly perception tests. This approach permits us to understand and define more precisely the role of F0 in the characterisation of Vietnamese tones.

Influence of F0 on Vietnamese Syllable Perception

Understanding and managing tonal characteristics of Vietnamese language is one of the most difficult aspects in Vietnamese speech processing. Our newest results indicate that, the initial consonant of one Vietnamese syllable does not carry information of the tone, the Vietnamese tone has an effect only on the Final part of the syllable. Based on obtained results, this article proposes linear F0 contour models for the generation of Vietnamese tones. These models only describe the F0 evolution of a final part of the Vietnamese syllable, and they are evaluated through perception tests.

Linear F0 Contour Model for Vietnamese Tones and Vietnamese Syllable Synthesis with TD-PSOLA

Being necessary for a Text-To-Speech (TTS) system, text-normalization is general a challenging problem, especially for Vietnamese because of the local context. Recent researches in text-normalization in Vietnamese for TTS systems are still at the beginning with very simple sets of ad hoc rules for individual cases in spite of the ambiguity of real text. The purpose of this paper is to take some initial steps towards methodically normalizing input text in Vietnamese for a TTS system. This paper proposes a categorization and a normalization model for Vietnamese text based on related results for other languages. An experimental application is implemented to demonstrate the model, which uses several techniques including letter language model and decision trees for classifying NSWs and both supervised and unsupervised approaches for expanding abbreviations.

A method for Vietnamese text normalization to improve the quality of speech synthesis

L'elaboration d'un systeme de synthese de la parole de haute qualite represente un travail considerable. Elle necessite des connaissances principalement dans deux domaines: la linguistique et le traitement du signal. A cause de la diversite intrinseque des langues, concevoir pour les systemes de synthese une structure universelle commune a toutes les langues semble encore illusoire. Les travaux de cette these non seulement choisissent, appliquent et evaluent des techniques et des theories developpees par ailleurs pour d'autres langues, mais encore contribuent a completer et a parfaire ces techniques et theories pour les adapter a la langue vietnamienne. Cette these se focalise egalement sur l'amelioration de la qualite du signal synthetique specifiquement pour le vietnamien par le biais d'une modelisation des parametres acoustiques et temporels de la prosodie de cette langue asiatique tonale, modelisation qui a notre connaissance n'a pas encore ete mise en oeuvre.

Do Dat Tran

Papers

Spoken and Written Language Resources for Vietnamese

Influence of F0 on Vietnamese Syllable Perception

Linear F0 Contour Model for Vietnamese Tones and Vietnamese Syllable Synthesis with TD-PSOLA

A method for Vietnamese text normalization to improve the quality of speech synthesis

Synthèse de la parole à partir du texte en langue vietnamienne