• Neha Shanbhag
• Shreyashi Samaddar
• Shashank Gajbhiye

Communication is a very important aspect of our day-to-day life, from chatting with our friends to national safety. Noise is an unavoidable factor in communication. Noise contaminates the characteristics of the speech signal and degrades its quality. This project identifies and removes non-speech components from speech signal. It was observed that the old techniques for noise suppression had some drawbacks such as when the energy of speech signal is lower than the threshold the speech signal is considered as noise and is subtracted, and the output is given. There is a need to design an algorithm which will not suppress the speech signal even when its energy is low. This project is set out to detect speech signal present in the audio signal contaminated by noise and remove non-speech components. The project starts with understanding the characteristics of an audio signal, performing basic operations such as sampling on the audio signal. These characteristics were obtained using Voice Activity Detection methodology. This in turn created a realization of voice identification in a speech signal. Further we take the live input and using our formulated algorithm we removed the silent parts from the input audio file.

silent part removal, ZCR(Zero crossing rate), PSD(Power Spectral Density)

"IDENTIFICATION AND REMOVAL OF NON-SPEECH REGIONS IN SPEECH SIGNALS", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.9, Issue 4, page no.i109-i128, April-2022, Available :http://www.jetir.org/papers/JETIR2204813.pdf

"IDENTIFICATION AND REMOVAL OF NON-SPEECH REGIONS IN SPEECH SIGNALS", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.9, Issue 4, page no. ppi109-i128, April-2022, Available at : http://www.jetir.org/papers/JETIR2204813.pdf