Hemanta Kumar Palo ¹, Jyoti Mohanty ¹, Mihir Narayan Mohanty ¹, Mahesh Chandra ²

Affiliations
1 Department of Electronics and Communication Engineering, Siksha ‘O’ Anusandhan University, Bhubaneswar - 751030, Odisha, IN
2 Department of Electronics and Communication Engineering, Birla Institute Technology, Ranchi - 835215, Jharkhand, IN

Abstract

Background/Objectives: The objective of this paper is to distinguish similar and overlapped emotional states like disgust and irritation from the primary angry state based on similarity measures. Methods/Statistical Analysis: The similarity test among these emotions have been carried out using the signal waveform, frequency coherence, power spectral density, log-likelihood score and Dynamic Time Wrapping (DTW) technique. Findings: Disgust state has more number of frequency coherence peaks which are leaning more towards unity like angry state. Further the phase angle between irritation and anger state is larger than between anger and disgust state. The minimum cost path using dynamic time wrapping technique founds to be 147.5917 between disgust and angry state as compared to 184.2386 between angry and irritation samples. From these analyses, it is concluded that, the primary anger state is more closure to disgust state than irritation state. The results are promising and we are able to put a boundary among these emotional states with our chosen techniques. Application/Improvements: Anger detection can lead to an improvement in human relationship and social system refinement. Demarcation of subcategory emotions that leads to anger state prevents confusion and can improve its recogntion.

Keywords

Emotional States, Dynamic Type Wrapping, Short Time Fourier Transform, Similarity Measure, Spectral Coherence.

Full Text

   

Rashmirekha Ram ¹, Mihir Narayan Mohanty ¹

Affiliations
1 Department of Electronics and Communication, ITER, Siksha ‘O’ Anusandhan University, Bhubaneswar - 751030,Odisha, IN

Abstract

Background/Objectives: The improvement of the speech quality is an essential component in this modern world. Though different methods have been used by many researchers since a long period, still the research field is open to work further. In this paper, we have analyzed the same using adaptive algorithms. Methods/Statistical Analysis: The algorithms like Least Mean Square (LMS) and Recursive Least Square (RLS) are used to improve the quality. Further, State-Space Recursive Least Square (SSRLS) algorithm is developed for enhancement of the noisy speech. The existing Spectral Subtraction (SS) algorithm is verified for two types of noisy speech. Also the standard adaptive algorithms have been experienced for these signals. Findings: Spectral Subtraction method is implemented initially. Then LMS and RLS algorithms are tested for two different types of noisy speech signals. In case of LMS, the improvement of SNR is 8.2822 dB for random noise and 6.5245 dB for babble noise. Whereas for RLS algorithm, the SNR improvement is 11.7152 dB for random noise and 14.3883dB for babble noise. The SNR after enhancement is 43.5409 dB and 47.2456 dB for random noise and for babble noise respectively in case of SSRLS algorithm which is much higher than other algorithms as results suggest. Also the error minimization curve converges more rapidly than the others. So SSRLS algorithm provides better enhanced signal than the other three methods. Application/Improvements: The applications are in different field of communications like mobile telephony, video conferences as audio and video applications. Also the applications are of speech recognition, voice communications, information forensics and many more. The improvement of SNR in this proposed algorithm found better than SS, LMS and RLS algorithms as 31.2377 dB and 35.6381 dB.

Keywords

Adaptive Filter, Least Mean Squares, Noise, Recursive Least Squares, Spectral Subtraction, Speech Enhancement,State-Space Recursive Least-Squares.

Full Text