Sinhala Text To Conversation System Development | Research

The system, that i am producing, called SINHALA Content material TO SPEECH is a one kind of fully research study. This paperwork briefly describes the operation of my STTS and highlights the top and benefits associated with the project. So this system allows user to go into sinhala texts and internally it will convert directly into pronunciation form. Actually it'll happen after user choose the particular option (convert to words) to convert it directly into that pronunciation form. So totally this technique is with the capacity of accepting individuals in sinhala words (sinhala fonts) and makes them in to sound waves, which can be captured by way of a technical object (speakers). Consumer will in a position to select the words type, which he/she like, it imply there are three option called child voice, female speech and adult (men) voice to select. By selecting that function individual can hear the tone, which he/she like the majority of. And the machine will carry out many perks to users, those who will utilize this system. The users who are not able to read sinhala, but those can understand verbally will encourage to use this system, because using this product they can conquer that problem quickly. If somebody needs documents with sinhala texts, then they might utilize this system to get that certain. In today world there are no such systems for sinhala language like that i will develop.

Table of Contents


Table of Articles 3



2. Purpose 5

3. Research PROBLEM 5


5. Books REVIEW 6


7. PROPOSED Methodology 8

7. 1 Consumer 8

7. 2 Data 8

7. 3 Suggestions 8

7. 4 Techniques 9

7. 5 Productivity 9


8. 1 Database Technology 9



9. 1. 1 Design Structures 10

9. 1. 2 Word process Architecture 11

9. 1. 3 Tone of voice Label Selection Process 12

9. 1. 4 Tone of voice Control Process 13


11. Bibliography 14

11. 1 SPEECH Research AND SYNTHESIS 14




"Sinhala Content material To Conversation" is the machine I hope to develop as my final research project. Like a post graduate college student I selected a research project that will convert the Sinhala input text message into a verbal form.

Actually, the term "Text-To-speech" (TTS) identifies the transformation of input word into a spoken utterance. The type is a Sinhala words, which may contain lots of words, sentences, paragraphs, figures and abbreviations. TTS engine should identify it without any ambiguity and make the corresponding speech sound influx with satisfactory quality. The result should be understandable for an average receiver without making much effort. This means that the productivity should be produced as close as to the natural speech quality.

Speech is produced when air is obligated from the lungs through the vocal cords (glottis) and along the vocal tract. Speech is split into a rapidly varying excitation signal and a slowly but surely varying filtration system. The envelope of the energy spectra provides the vocal tract information.

The verbal form of in insight should be understandable for the receiver. This means that the outcome will be made as nearer as the natural real human voice. My system will perform few main features. A few of them are, after stepping into the text consumer will capable of selecting one of tone features, means women speech, male voice and child tone of voice. Also an individual is capable of doing variation in quickness of the voice.

Actually, my job will perform main few advantages to the users, those who intend to use this.

Below I've mentioned the essential architecture of our own project.

Sinhala Voice

Text in Sinhala


Voice and speed



Figure 1. 1


To create a system, that can able to read content material in sinhala format and covert it directly into verbal (sinhala) form. And also, It will capable to change the acoustics waves, It signify user would able to select tone of voice quality corresponding to his/her point of view. You will find might be three tone selections. They are kind of woman words, kind of male tone and kind of kid's voice. And user can change the rate of the words. If somebody must hear low velocity voices or high-speed tone of voice, then he/she can change it according with their requirements.


Actually before start this task I've accessed into the Internet and collect more information regarding this specific field. First-of-all I have to provide a center to type in sinhala font into the computer. So, to triumph over this subject I intend to use UNICODE. Whenever we pronounce sinhala words, sometime we need use pronouncing voices of two texts.

It means to create voice for some texts we have to combine another two wording voices. In like manner have voices we ought to store voices to every single content material in the tone of voice data source. Then voices result from voice database in line with the words which we joined. Actually directly after we entered wording internally it (texts) get in to different groupings.


The thought of developing a Sinhala Content material To Talk (STTS) engine unit have begun once i taking into consideration the opportunities available for Sinhala speaking users to grasp the benefit for Information and Computer Technology (ICT). In Sri Lanka more than 75% of inhabitants talks in Sinhala, but it is rather unusual to find Sinhala softwares or Sinhala materials regarding ICT in market. This is immediately effect to development of ICT in Sri Lanka.

In present few Sinhala words to talk softwares can be found but those have problems such as quality of sound, font schemas, pronunciation etc. Because of these problems developers are afraid to utilize those STTS because of their applications. My give attention to developing an engine that can convert Sinhala words in digitized form to Sinhala pronunciation with error free manner. This engine will develop some applications.

Some applications where STTS can be used

Document reader. An already digitized report (i. e. e-mails, e-books, newspapers, etc. ) or a typical document by scanned and produced through an optical persona recognizer (OCR).

Aid to handicap person. The eyesight or tone impaired community may use the personal computers aided devices, directly to communicate with the entire world. The vision-impaired person can be up to date by a STTS system. The voice-impaired person can communicate with others by giving a keypad and a STTS system.

Talking books & playthings. Producing talking literature & toys and games will boost the playthings market and education.

Help assistant. Develop help assistant talks in Sinhala like in MS Office help assistant.

Automated News casting. The future of entirely new variety of television networks that contain programs managed by computer-generated characters is possible.

Sinhala Text reader. SMS consist of several abbreviations. If something that read those announcements it can help to receivers.

Language education. A high quality TTS system incorporated with a computer-aided device can be utilized as a tool, in learning a fresh dialect. These tools can help the learner to improve very quickly since he/she has the access to the right pronunciation whenever needed.

Travelers guide. System that located inside the automobile or mobile device that gives information current location & other relevant information incorporated with GPRS.

Alert systems. Systems that can be incorporated with a TTS system to appeal to the attention of the managed elements since as humans are being used to sketch attention through speech.

Specially, countries like Sri Lanaka, which continues to be battling to harvest the ICT benefits, may use a Sinhala TTS engine unit as a remedy to convey the information effectively. Users

can get required information after that native terms (i. e. by converting the text to native words wording) would obviously move there thoughts to the attainable benefits and you will be encouraged to use information technology much frequently.

Therefore the introduction of a TTS engine motor for Sinhala provides personal benefits (e. g. help for handicapped, dialect learning) in a social perspective and definitely a financial benefit in economical terms (e. g. virtual tv networks, toys manufacture) for the users.


"Wording to speech "is extremely popular area in computer science field. There are many research held upon this area. Most of research basic on "how to build up more natural conversation for given word ". You can find freely available text to speech package available on the planet. But almost all of software develops for most common vocabulary like English, Japanese, Chinese languages. Even some software companies spread "text to speech development tools "for English dialect as well. "Microsoft Conversation SDK tool set up" is one of the examples for freely distributed tool kit produced by Microsoft for English language.

Nowadays, some universities and research labs doing their research study on "Word to speech". Carnegie Mellon School performed their research focus on text to talk (TTS). They provide Open Source Talk Software, Tool kits, related publication and important ways to undergraduate student and software programmer as well. TCTS Lab also doing their research on this area. They introduced simple, but general functional diagram of the TTS system [Ref. 2].

Image Credit: Thierry Dutoit.

Figure5. 1. A straightforward, but general efficient diagram


Produce a verbal format for the suggestions sinhala content material.

Input Sinhala wording which may be a user type or a given text report will be changed in to sound waves, which is then output is captured by sound system. Therefore the disabled people will be one of the most beneficial stakeholders of Sinhala Text to Speech system. Also undergraduates and research people who need to make use of more referrals can send the written text to my system, just pay attention and get what they need.

The outcome would become more like natural talk.

The human voice is a intricate acoustic indication, which is produced by an air stream expelled at either mouth area, nasal or both. Important characteristics of the speech sound are rate, silence, accentuation and the amount of energy result. The tongue correctly controls air steam, lips with the aid of other articulators in the vocal system. Many versions of the speech signal are caused by the person's vocal system, to be able to convey the meaning and feelings to the receiver who then understand the note. Also includes a great many other characteristics, that are in receiver's ability to hear system to identify what is being said.

Identify an efficient way of translating sinhala content material in to verbal form.

By developing this system we would be able to identify and suggested a the most suitable algorithm, which may be used to translate sinhala format to verbal form by an easy and successful manner.

Control the tone of voice speed and types of the speech (e. g. man, women, child speech, etc. ).

Users would be capable of selecting the quality of the sound influx, which they want. Also they would be allowing to reset the swiftness of the result as they need. People, those would like to learn Sinhala as their second vocabulary to learn elocution properly by changing the swiftness (lowering and increasing). Which means this will enhance the listening capabilities.

Small kids can be motivated to learn terms by differing the velocity and types.

Propose ways for that may be extended the existing system further more for future needs.

This system only provides basic functions. My system is possible of enhancing further more in order to gratify the changing requirements of the users. This is embedded directly into toys so may be used to improve children being attentive and elocution ability. So those will Borden their speaking capacity.


Main function of my system is read sinhala digitized people and speak out those words as closer noises that real human natural words.

7. 1 User

My basic idea is to build up systems that provide all types of users. Which means that who know the operate computer perfectly and also who's beginner to the computer field. Users only wish to accomplish insert text message in sinhala.

7. 2 Data

In my database I am hoping to store speech tags, sinhala individuals and pronunciation rules. And also I wish to introduce efficient algorithms for search relevant speech tags from the database.

7. 3 Input

Proposed system will get sinhala-digitized characters, speech selection as source.

7. 4 Processes

Get the phrase from the user and it'll recognized end of word by full-stop and it'll divide two words by the space between two words. Those words will break down to smaller parts. Then after get the relevant voice tags regarding to rules that I've given and combine those tone tags. Then after get words selections that end user given and process to provide those sound files.

7. 5 Output

Produce the related sinhala voices for content material that is distributed by the user matching to sinhala pronunciation rules as well as speech selection done by an individual.


8. 1 Database Technology

Hope to make use of OO methodologies and Relational Database Management System (Microsoft SQL Server 2005) to build up centralized database on main server. A database management system, or DBMS, is software design to assist in retaining and utilizing large assortment of data [Ref. 3]. The SQL Server 2005 is design to are a data storage space engine unit for thousand of concurrent users who hook up over the network, additionally it is with the capacity of working as a stand-along database directly on the same computer as a credit card applicatoin [Ref. 4]. DBMS provide some important functionality. Applications are unbiased from data representation, storage space and location (data and location independence). DBMS can check out through million of record and get efficiently (productive data gain access to). DBMS enforce integrity constrain and security permission on the info (data integrity and security). DBMS provide facilities to data and its own efficient availability (data administration). DBMS routine concurrent usage of the info in such manner that individual can think of the info as being accessed by one user at a time. Further, DBMS protects users from the effects on of system failures (concurrent gain access to and crash recovery). There for hope to use Microsoft SQL Server 2005 to build up voice and words information database.



9. 1. 1 Design Architecture

Text in Sinhala

Voice and swiftness selection


Sinhala voice

Figure 9. 1. 1

Speed selection

Array of word (Sinhala)

Process in detail

Related Sinhala




Process the Text

Get the voice tags according to the Text and combine them

Voice selection

Voice controller

Figure 9. 1. 2

Figure 9. 29. 1. 2 Text message process Architecture

Detect full-stops, commas, brackets etc.

Separate out numbers

Get unique number to each notice and store it in an array

Send the info in array to voice tag selection process

Separate the written text to sentences

Group the written text regarding to letters

Sinhala Text

Array of letter values

Figure 9. 1. 3

This process gets a words as the insight. It find whether there are any full-stops, commas etc. to avoid confusions. If there any figures in the text they are split out and text message is partition in to sentences. After that each notice in a phrase grouped, provide a unique number store within an array. This array is send to the next process.

9. 1. 3 Speech Tag Selection Process

Figure 9. 1. 4

Voice Database

Get words tags from tone of voice Database

Voice selection

Array of notice values

Merged Tone tag

Merge tone tags to the order

Send the merged words tags to tone Control process

Select the voice type

This process gets the array, which gives from Wording process and speech selection as inputs. By using these inputs this process gets speech tags for each letter and combine them. Merge tone tags send to the tone of voice control process.

9. 1. 4 Tone Control Process

Figure 9. 1. 5

Speed selection

Sinhala Voice

Store the voice text array

Control the rate Voice speed

Speak the speech array

Merged Speech tag

This process gets merge voice tags and tone of voice rate selection as insight. It set up the Merge speech tags regarding to speed decided on. Then it will speak out conversation each voice tag.


[Ref. 1] Building Synthetic Voices, [Online] http://www. festvox. org/festvox/

[Ref. 2] An Introduction to Text-to-Speech Synthesis, [Online] http://tcts. fpms. ac. be/synthesis/introtts. html

[Ref. 3] - Raghu Ramakrishnan, Johannes Gehrke/Database Management System

Third release - 2001/ McGraw-Hill

[Ref. 4] - SQL Server Catalogs, [Online] 1988-2005 Microsoft Organization.

Also We Can Offer!

Other services that we offer

If you don’t see the necessary subject, paper type, or topic in our list of available services and examples, don’t worry! We have a number of other academic disciplines to suit the needs of anyone who visits this website looking for help.

How to ...

We made your life easier with putting together a big number of articles and guidelines on how to plan and write different types of assignments (Essay, Research Paper, Dissertation etc)