<
>
+1 -1 → +13
Vote on this proposal

Speech Recognization

by Anshul Gupta (speaking)

Section
Scientific Computing
Technical level
Intermediate

Objective

In this session we will be discussing to develop a speech commanding software system for the Linux environment.Will be using modules like matlibplot ,scipy ,numpy . Speech recolonization system works on the trained voices stored in the system . It finds the best possible match from the database and then follow desired steps to full fill that command. this will give user freedom to operate system more convenient for some distance.

Description

Speech recognition is the process of enabling a computer to identify and respond to the sounds produced by human speech. It basically come under three broad categories that are 1.acoustic phonetic
approach ,which tries directly exploit individual sound properties. 2. Pattern recognition approach which realize only on gross spectral and temporal properties of speech sound . 3. Artificial approach , involves
learning.
Existing systems do not have much accuracy to recognize human voice. We will try to increase the accuracy of speech recognition system in compared to existing system. Speech recognition system createsthe acoustic input as if it were a "noisy" version of the source command. In order to decode this noisy command , we will consider all possible commands and for each one we compute the probability of it generating the noisy word. It will also consist of additional Facilities such as file reading , weather telling.

Requirements

Basic knowledge of linux , python.

Speaker bio

I am an engineering student from computer science branch batch 2011-2015.I do software development in python .I made software for cricket team selection for my university using python. I know core ,GUI development , network programming. kivy for android in python

Comments

Login with Twitter or Google to leave a comment →