IEEE paper and Plagiarism Report

Topic: Speech Recognition Using Correlation

This paper was made by:

 Parth Panchal,Soumyaa Passari ,Pranali Patil,Purvika Patil and Suma Salian

Attached below is the Link for IEEE Paper and Plagiarism Report

https://drive.google.com/drive/folders/0BwkjNOLlpFQNek9jYXQzTENxY1U?usp=sharing

Patent Review

Patent : Speech recognition circuit using parallel processors

USPTO Applicaton #: #20170110122 
Inventors: Mark Catchpole

         A speech recognition circuit comprises an input buffer for receiving processed speech parameters. A lexical memory contains lexical data for word recognition.Conventional large vocabulary speech recognition can be divided into two processes: front end processing to generate processed speech parameters such as feature vectors,followed by a search process which attempts to find the most likely set of words spoken from a given vocabulary (lexicon).The front end processing generally represents no problem for current processing systems.

         This aspect of the present invention provides a circuit in which speech recognition processing is performed in parallel by groups of processors operating in parallel in which each group accesses a common memory of lexical data. This aspect of the present invention provides the advantage of parallel processing of speech parameters and benefits from a limited segmentation of the lexical data. By providing a plurality of processors in a group with a common memory, flexibility in the processing is provided without being bandwidth limited by the interface to the memory that would occur if only a single memory were used for all processors. The arrangement is more flexible than the parallel processing arrangement in which each processor only has access to its own local memory and requires fewer memory interfaces (i.e. chip pins). Each processor within a group can access the same lexical data as any other processor in the group.

IEEE Paper Review

Paper Title:Brain-Computer Interface Technology for Speech Recognition

Authors: Mashael M. AlSaleh Mahnaz Arvaneh Heidi Christensen and Roger K. Moore

This paper presents an overview of the studies that have been conducted with the purpose of understanding the use of brain signals as input to a speech recogniser. the paper gives an insight into some studies that examined the effect of the chosen stimuli on brain activities as an important factor in the recognition process. The remaining part of this paper lists the limitations of the available studies and the challenges for future work in this area.
Brain Computer Interface (BCI) is one of the promising   technologies that has been examined as an alternative communication technology. Neuro muscular impairments prevent users from using most of the available communication aids, since they require some degree of muscle movement. This makes them unsuitable for people with severe disabilities who have limited movement in their muscles, such as people with locked-in syndrome. In more general circumstances it would be desirable to communicate only using brain activities,e.g., due to security issues.
In the literature, brain activity has been used for communication in two different ways: controlling spellers and capturing speech information. This review paper focuses on BCI studies related to speech. These studies mainly focus on the following objectives: a) understanding the
mechanism of spoken (i.e. overt) and unspoken (i.e. covert) speech production in the brain, b) recognising speech from covert speech using brain signals.This review covers two main parts. 

Perform basic operations using DSP processor

 In this experiment ,we used DSP kit TMS320F28375  to perform basic mathematical operations also Code Composer Studio Software was used.Arithmetic instructions like addition, subtraction and multiplication were performed. Logical instructions like AND and NOT were performed. Also, shift operations like logical shift left , shift right, rotate left and rotate right were performed.

FIR filter design using FSM

In this experiment, we designed a FIR filter using the Frequency Sampling Method(FSM). The filter was designed by coding in C language in Scilab. .In this technique the desired frequency response is uniformly sampled and inverse DFT is performed to obtain the corresponding impulse response
.Magnitude and phase response were plotted for observation and verification.

FIR filter design using Window function

FIR is Finite Impulse Response.In this experiment,we designed a linear phase FIR filter using scilab software and implemented. The attenuation in passband and stopband are provided as input and the magnitude response for high pass and low pass filter is plotted. The window function is selected depending on the value of attenuation in stop band.We used Hanning Window function in this experiment.

Experiment 6 : Chebyshev filter design

        In this experiment , we designed and implemented Chebyshev filter on Scilab software.This was similar to previous experiment on Butterworth filter design. Chebyshev filters have the property that they minimize the error between the idealized and the actual filter characteristic over the range of the filter,but with ripples in the passband.
         Chebyshev filter is designed using necessary parameters and its transfer function is calculated. Magnitude spectrum  is  plotted to observe passband and stopband . we observed that no of ripples were present in pass band.