An Automatic Daisy Production Tool for Thai Language to Support Production of Hand book for Disaster Preparedness in Thailand Wantanee Phantachat, Namnueng Mitsamarn, Waragorn Gestubtim, Sirilak Junnatas
National Electronics and Computer Technology Center National Science and Technology Development Agency Ministry of Science and Technology Pathumthani, Thailand Info-reat@nectec.or.th
Abstract
Thailand had faced many disasters around the year. The television news always showed floods, flash floods, landslides, mud slides, wind storm, forest fire and earthquakes in almost every regions of Thailand. Mae Hong son, Prae, Nan and Utaradit provinces are good examples of provinces that were regular hit by the depression and caused mudslides. Many villagers had got direct impacted from these disaster, some died, some injured, some were homeless. For disability concerned, we talked to the head of villages regarding the evacuate plan for people with disability (pwd) in the village. He said he didn’t know how to help them because it happened instantly. Everyone had to run for their lives. With respect to the information, we realize that to train pwd to learn how to prepare themselve for any disaster is very important. Due to disability concerned, the training guides should be prepared to support all types of pwd. In this paper we propose the production tool that can produce the handbooks for disaster preparedness in Thai language automatically. We use ready to made technology and try to integrate them into one program that can produce the handbooks in daisy format. The “Save as Daisy “ program from Microsoft is our crucial main program combining with NECTEC speech syntheses program , NECTEC optical character recognition program, NECTEC Thai word segmentation engines, NECTEC automatic Thai typing style checking engines, and Thai styles sheet are integrated into one program. In this paper is also presented how to produce easy Daisy handbook for disaster preparedness.
Rational
The statistical numbers of disasters in Thailand are growing every year. Mostly Thailand faces floods, flash floods, landslides, mud slides, wind storm, forest fire and earthquakes in every regions of Thailand. The television news reported these disasters caused many people death, injured, home damaged, farm damaged rice field damaged etc. For disability concerned, we talked to the head of villages regarding the evacuate plan for people with disability (pwd) in the village when disasters happened. He said he didn’t know how to help them because they happened instantly. Everyone had to run for their lives. With respect to the information, we realized that to give opportunity to pwd to access to public knowledge on disaster preparedness is very important. Including Inclusive evacuate plan should be considered in every villages for all villagers who live in the villages. The people who are left out ; the blind, the people with physical disability, the people with intellectual disability, people with autistic, the deaf, the hill tribe people and other minority race people can access to the knowledge by using their own language and any form of media they preferred.
Objectives
To make the public knowledge on disaster preparedness available for everybody, the accessible information should be considered to be the media for communication. The Digital Accessible Information System Specification is one of the appropriate media form for using to produce knowledge distribution. The output of the media can be in the form of text, voice or pictures or multimedia form of text, voice or pictures which are synchronized. The readers of the media can see what they hear. This is to help the all readers who are printed disability (can’t read the printed material). The technology allow the reader to listen to the text by either syntheses voices or recorded voices. So the people who are left out as stated above can all access to the media according to what language they use, what media form they like.
Many years ago, there are a few tools to produce such media. It is said that to produce the accessible media or Daisy book is time consuming process. The amount of daisy books produced in each country are not as much as expected. So in this paper we proposed an simple production tool that can produce the handbooks for disaster preparedness in Thai language automatically. We use ready to made technology and try to integrate them into one program that can produce the handbook in daisy format. The “Save as Daisy “ program from Microsoft is our crucial main program combining with NECTEC speech syntheses engine, NECTEC optical character recognition engine, NECTEC Thai word segmentation engines, NECTEC automatic Thai typing style checking engines, and Thai styles sheet are integrated into one program.
Conceptual Model Diagram of the Daisy Production Tool
We propose the conceptual model diagram for Daisy Production tool as follows:
Diagram 1: conceptual model diagram for a Daisy Production tool
There are 3 modules in this model diagram. The first module is a pre-production module. It is for transfer printed material to text file using Thai Optical Character Recognition Program, then using Text Proof Reading Program to do word autocorrection and Thai typing regulation checker of text output from OCR program. The second module is adapted “Save as Daisy Program” using Thai style sheet format. In this module we also add in NECTEC Speech Synthesis engine as one of the alternative speech output. The last module is Daisy editor. This module is proposed to add in as the last module for editing the Daisy book format.
With this model we expect that the daisy production tool can save the time to produce a book 5 time less than the traditional tool.
Examples of Disaster handbook Production.
With this tool, we hope that all public information can be able to transfer into Daisy book more easily. The left out people will have opportunity to access to knowledge as same as the others.
In this paper , we propose the 5 steps for making simple daisy book as follows:
1. Transfer the printed paper to text file and run auto proof reading program.
2. Mark up text into Daisy Style
e.g. Top Level markup to Heading Level1
e.g. Top Level markup to Heading Level1
3. Save document as file.docx
4. Validate the file according to DAISY format
5. Choose Save as DAISY. Then select speech synthesis : Open TTS –> SaveAsDAISY -> Full DAISY -> DAISY Pipeline
Select output directory : Edit document properties -> Title, Creator, Publisher .-> Translate
Conclusion
The simple Daisy Production tool will open the accessibility world for many people who intend to share their own information in the accessible format. No one will excuse for unaccessible file when they have to share file to others to read. We hope that with this tool, daisy community network will piled up with the accessible knowledge that anyone can contribute for those who seeks for information.
Bibliography:
Daisy Consortium, Digital Accessible Information System specification (ANSI/NISO Z39.86 Specifications ) .http://www.niso.org/workrooms/daisy/Z39-86-2005.html
Department of Disaster Prevention Mitigation , Ministry of Interior, 2009. .Preparation and Code of Practice when Floods, Wind Storm and Mud Slide Occur.
Source Forge.net. “The “Save as DAISY XML “.
http://sourceforge.net/project/platformdownload.php?group_id=208610
Acknowledgement:
The “Save as DAISY XML” add-in was created through an open source project with Microsoft, Sonata Software Ltd. and the Digital Accessible Information System (DAISY) Consortium
NECTEC Speech Syntheses is developed by Human Languguage Technology laboratory ,NECTEC
NECTEC Optical Character Recognition is developed by Image Technology Laboratory, NECTEC
NECTEC Proof reading Program is developed by Rehabilitation Engineering and Assistive Technology Institute and Human Language Technology laboratory, NECTEC