vi
en
Neural Text Normalization in Speech-to-Text Systems with Rich Features
Content
Authors: Oanh Thi Tran, Viet The Bui
Abstract: This paper presents the task of normalizing Vietnamese transcribed texts in Speech-to-Text (STT) systems. The main purpose is to develop a text normalizer that automatically converts proper nouns and another context-specific formatting of the transcription such as dates, time, and numbers into their appropriate expressions. To this end, we propose a solution that exploits deep neural networks with rich features followed by manually designed rules to recognize and then convert these text sequences. We also introduce a new corpus of 13 K spoken sentences to facilitate the process of text normalization. The experimental results on this corpus are quite promising. The proposed method yields 90.67% in the F1 score in recognizing sequences of texts that need converting. We hope that this initial work will inspire other follow-up research on this important but unexplored problem.
Published: 11 Jan 2021
DOI: 10.1080 / 08839514.2020.1842108
This research is funded by International School, Vietnam National University, Hanoi (VNU-IS) under project number CS.NNC/2020-05.
Do you need a workthrough of our platform? Let us know
Do you need a workthrough of our platform? Let us know
Related Posts
Operate an Intelligent Call Center with Conversational AI
Contact center remains a “priority investment” for businesses to quickly and effectively serve customers. Artificial Intelligence (AI) has proven itself to be the key solution to sophisticated business problems, in which 79% of call centers expect to invest in this technology (Deloitte, 2021). Prominently, Conversational AI solution will help automate two-way interaction with customers, allowing … Continued
Multifeature Image Indexing for Robot Localization in Textureless Environments
Content Authors: Tran Duc Dung; Delowar Hossain; Shin-ichiro Kaneko; Genci Kapi Abstract: Robot localization is an important task for mobile robot navigation. There are many methods focused on this issue. Some methods are implemented in indoor and outdoor environments. However, robot localization in textureless environments is still a challenging task. This is because in these environments, the scene … Continued
Neural Network based 3D Mapping Using Depth Image Camera
Content Authors: Tran Duc Dung; Genci Capi Comments: Accepted at the Conference PACLIC 2020 Abstract: Mapping is a crucial task for robot navigation. Especially, in order to develop a fully autonomous robot that can interact well with human, the map is not only required to contain geometry but also the semantic contents. Building a detailed map of the … Continued
A Faster R-CNN Approach for Partially Occluded Robot Object Recognition
Content Authors: Delowar Hossain, Sivapong Nilwong, Duc Dung Tran, Genci Capi Abstract: Many objects in household and industrial environments are commonly found partially occluded. In this paper, we address the problem of recognizing objects for use in partially occluded object recognition. To enable the use of more expensive features and classifiers, a region proposal network (RPN) which … Continued
Towards Understanding User Requests in AI Bots
Content Authors: Luong Chi Tho, Tran Thi Oanh Abstract: This paper presents the task of deeply analyzing user requests: the situation in ordering bots where users input an utterance, the bots would hopefully extract its full product descriptions and then parse them to recognize each product information (PI). This information is useful to help bots better understand … Continued
Understanding what the users say in chatbots: A case study for the Vietnamese language
ContentAuthors: Luong Chi Tho, Tran Thi Oanh Abstract: This paper1 presents a study on understanding what the users say in chatbot systems: the situation where users input utterances bots would hopefully detect intents and recognize corresponding contexts implied by utterances. This helps bots better understand what users are saying, and act upon a much wider … Continued
A combined syntactic-semantic embedding model based on lexicalized tree-adjoining grammar
ContentAuthors: Le Hong Phuong, Dang Hoang Vu Abstract: This paper presents a joint syntactic-semantic embedding model which not only uses syntactic information to enrich the word embeddings but also generates distributed representations for the syntactic structures themselves. The syntactic input to our model comes from a Lexicalized Tree-Adjoining Grammar parser. The word embeddings from our … Continued
An Efficient Algorithm for the k-Dominating Set Problem on Very Large-Scale Networks (Extended Abstract)
Content Authors: Tran The Trung, Nguyen Minh Hai, Ha Minh Hoang, Hoang Thai Dinh, Eryk Dutkiewicz, Diep N. Nguyen Abstract: The minimum dominating set problem (MDSP) aims to construct the minimum-size subset D⊂VD⊂V of a graph G=(V,E)G=(V,E) such that every vertex has at least one neighbor in D. The problem is proved to be NP-hard. In a recent industrial application, we encountered a more general variant … Continued
Introducing a Large-Scale Dataset for Vietnamese POS Tagging on Conversational Texts
Content Authors: Oanh Tran, Tu Pham, Vu Dang, Bang Nguyen Abstract: This paper introduces a large-scale human-labeled dataset for the Vietnamese POS tagging task on conversational texts. To this end, wepropose a new tagging scheme (with 36 POS tags) consisting of exclusive tags for special phenomena of conversational words, developthe annotation guideline and manually annotate 16.310K sentences … Continued
Online Adaptation of Language Models for Speech Recognition
Content Authors: Dang Hoang Vu, Van Huy Nguyen, Phuong Le-Hong Abstract: Hybrid models of speech recognition combine a neural acoustic model with a language model, which rescores the output of the acoustic model to find the most linguistically likely transcript. Consequently the language model is of key importance in both open and domain specific speech recognition and … Continued
Towards Task-Oriented Dialogue in Mixed Domains
Content Authors: Luong Chi Tho, Le Hong Phuong Abstract: This work investigates the task-oriented dialogue problem in mixed-domain settings. We study the effect of alternating between different domains in sequences of dialogue turns using two related state-of-the-art dialogue systems. We first show that a specialized state tracking component in multiple domains plays an important role and gives … Continued
Building Vietnamese Linguistic Resources for Social Network Text Analysis
Content Authors: The-Tuyen Nguyen; Xuan-Luong Vu; Phuong Le-Hong Abstract: In this paper, we report our work on building linguistic resources for Vietnamese social network text analysis in multiple domains. We first describe our annotation methodology including guidelines development, annotation softwares and quality assurance. We then present results of the first pilot phase of the project. Finally, we outline some … Continued
Get ahead with AI-powered technology updates!
Subscribe now to our newsletter for exclusive insights, expert analysis, and cutting-edge developments delivered straight to your inbox!