Design of recognize the words “Yes” and “No” played from a sound using MATLAB

Introduction

MATLABSolutions demonstrate In this task we are going to design In this project we are going to conduct the digital signal processing analysis on the audio files with the word yes and no sound with and without sinusoidal wave. We first gather all the audio files with yes and no sound after that we will create the algorithm which will detect the words in audio using the spectrum analysis. Analysing audio recordings with the words "yes" and "no" using digital signal processing (DSP) can provide important details about the properties of these sounds and their underlying frequency components. These sounds may have their frequency content further altered by the addition of a sinusoidal wave, enabling a more in-depth examination of the effects of various frequency components on the sound. These audio files can be subjected to DSP analysis using a variety of methods, including FFT, digital filtering, and time-frequency analysis. To identify and analyse individual frequency components, the audio signal can be transformed from the time domain to the frequency domain using FFT. Specific frequency ranges can be eliminated or reduced via digital filtering, which gives a sharper image of the remaining frequency components. The frequency content of the audio stream can be visually represented across time using time-frequency analysis techniques like spectrograms or wavelet transformations.

We will utilize MATLAB to create the algorithm and audio analysis, it is feasible to determine the effect of these extra frequency components on the overall sound by comparing the analysis of these audio files with and without sinusoidal waves. Understanding the impact of various forms of distortion or noise on audio signals can be very helpful in this regard. The distinctions between the "yes" and "no" sounds, such as variations in pitch or frequency content, may also be understood by contrasting the characteristics of the two sounds. The proposed technique is a fundamental example of how digital signal processing analysis on audio files may be carried out in MATLAB. The Fast Fourier Transform, digital filtering, and time-frequency analysis are the three primary DSP methods that are the emphasis of the program. These methods are frequently employed for the analysis of audio signals and can reveal information about the frequency content and properties of the signals. The analysis, improvement, and compression of audio and voice signals make substantial use of DSP methods. For instance, speech coding methods are used to compress speech signals for effective transmission across communication networks, digital filters are used to eliminate noise from audio signals, and FFT is used to examine the frequency content of audio signals.

Methodology

We have given the dataset of audio containing Yes and No word sound, first we have to plot the frequency response and spectrum of both the audio words after that we have to determine the cut off frequency for Low pass filter and high pass filter based on the frequency data by just visualizing, we can identify the difference between the magnitude, the yes sound contain more magnitude as it contains large amount of high frequency as compare to the No sound. Now we will design and apply the filter to the cut off frequency, to remove the sine wave we can design the filter which will remove the frequency of 1kHz and perform the fft of the signal which store the magnitude of all the audio signal separately for yes and no word. We will use the two-signal separated using LP and HP filter output and compute the Power ratio between those which will give the information about yes and no word. We will find the threshold value for the power ration for the no sin signal data and test the data algorithm the data which have contain noise and sin waveform as well which will be filtered out and the predicted results with actual class will be displayed on the command window. The FFT technique is frequently used for audio analysis in a variety of industries, such as music, speech, and signal processing. The Fourier transform, a mathematical method that separates a time-domain signal into its frequency components, is the foundation of the FFT theory.