• M.HIMA BINDU
• CVP Supradeepthi

Emotion recognition is one of the most active areas of research in the field of artificial intelligence. While emotions can be drawn from many modes like semantics, visual cues and even multimodal applications, speech alone carries a dominant part of emotions. To accomplish this task, many approaches have been formulated and applied over the years like Advanced signal processing, Machine learning and artificial neural networks. Although they are very reliable, they require a lot of resources in terms of training and testing which in turn increases the computational time and necessary hardware requirements. In this , a simple speech emotion recognition system was designed using a comparatively small data set and a simple deep neural network. An MFCC feature extractor was used to capture necessary pitch variations from the audio files and then these features were fed into a 1D convolutional neural network. This framework was further optimized by varying hyperparameters from both the feature extractor and the neural network which resulted in a decent accuracy over a less computational time and hardware.

CNN, Deep neural networks, MFCC, Feature extraction.

"DESIGN OF EFFICIENT SPEECH EMOTION RECOGNITION SYSTEM", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.9, Issue 6, page no.j742-j747, June-2022, Available :http://www.jetir.org/papers/JETIR2206993.pdf

"DESIGN OF EFFICIENT SPEECH EMOTION RECOGNITION SYSTEM", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.9, Issue 6, page no. ppj742-j747, June-2022, Available at : http://www.jetir.org/papers/JETIR2206993.pdf