User:Tachyons/GSoC-speech recognition

Developing Acoustic and Language Model for Malayalam Recognition

Personal information

Email Address       : aboockervyd@gmail.com
Blog URL            : abvayad.wordpress.com
Freenode IRC Nick   : tachyons
Your university and current education: CUSAT , B Tech Computer science

Why do you want to work with the Swathanthra Malayalam Computing?

I heard about smc by accident . I saw the mailing list from launchapad when I was helping one project for malayalam translation. I joined in that mailing list and really surprised with dedication of smc members for malayalam language computing. I asked to one of the members about the reason behind this dedication . His answer was simple "I am a malayalee ,It is our responsibility" . As a malayalee I also have responsibility to complete the dream "എന്റെ കമ്പ്യൂട്ടറിനു് എന്റെ ഭാഷ". So I decided to contribute to smc in this summer . I will work for smc even if I am not selected for GSoc , Because It is my responsibility

Do you have any past involvement with the Swathanthra Malayalam Computing or another open source project as a contributor?

I don't have considerable contribution to opensource projects .But I hope I can involve in open source projects in future

here is my github profile tachyons

Did you participate with the past GSoC programs, if so which years, which organizations?

No

Do you have other obligations between May and August ? Please note that we expect the Summer of Code to be a full time, 40 hour a week commitment

NO , I don't have such obligations

Will you continue contributing/ supporting the Swathanthra Malayalam Computing after the GSoC 2013 program, if yes, which area(s), you are interested in?

Yes I will , My interested fields are

• Machine language translation system for malayalam
• Malayalam Speech Recognition
• Malayalam text to speech(Improving Dhani)

Why should we choose you over other applicants?

I am a foss enthusiast and technology enthusiast. I am new to the field of digital sound processing ,But I have knowledge about digital signal processing and have solid background in major programming languages . I also planned to create a similar project as the academic project , So that I can spend more time and effort for this project

Proposal Description

Problem Statement

Malayalam (മലയാളം), is a language spoken in India, predominantly in the state of Kerala. It is one of the 22 scheduled languages of India. Malayalam has official language status in the state of Kerala and in the union territories of Lakshadweep and Puducherry. It belongs to the Dravidian family of languages, and is spoken approximately by 33 million people according to the 2001 census.

Malayalam do not have existing speech processing systems ,So in order to develop a speech processing sysstem we have to create an an acoustic model and language Model has to be developed for malayalam . This project isto create an acoustic model and language Model for malayalam so that it will be helpful for future development in malayalam speech Recognition system

Introduction

Speech is the ancient but still more powerfull way of sharing of information . Even an illiterate person can share information via speech . But unlike other communication ways like text , Speech processing is very difficult to achive . Due to this complexity there is no perfect speech Recognition tools are available today . Carnegie Mellon University created an open source tool called CMU sphnix , It made considerable improvement in speech Recognition

• Speech Recognition in CMU sphnix

In CMU sphnix common way to recognize speech is the following: Take the wave form and split it on the utterance , Process that slice and determine match it with predefined database .

• Models

Model describes some mathematical object that gathers common attributes of the spoken word.

• Acoustic models The acoustic model establishes a mapping between phonemes and their possible acoustic manifestations . It characterize how sound changes over time. It captures the characteristics of basic recognition units
• The language model describes the likelihood, probability, or penalty taken when a sequence or collection of words is seen. It attempts to convey behavior of the language and tries to predict the occurrence of specific word sequences possible in the language.

The need you believe it fulfills

• Acoustic model
• The language model

How and who it will benefit in society

Previous attempts

Challenges

References

Any relevant experience you have

How you intend to implement your proposal

A rough time-line for your progress with phases

Any other details you feel we should consider

Tell us about something you have created.

Have you communicated with a potential mentor? If so, who?

Not yet

SMC Wiki link of your proposal

http://wiki.smc.org.in/User:Tachyons/GSoC-speech_recognition