This inventor holds 1 USPTO granted patent. Top assignee: Tensortype Inc.. Active years: 2026.
Data source: USPTO (United States Patent and Trademark Office) public patent records. Weekly synchronization. How IDiyas builds this profile
Company Filing History:
Years Active: 2026
Title: Kranti Kumar Parida: Inventor in Audio-Visual Speech Denoising
Introduction
Kranti Kumar Parida is an inventor based in Bhubaneswar, India, and works at Tensortype Inc. His work includes a patent focused on reducing noise in audio-visual speech.
Latest Patents
Parida’s patent, “Methods and Systems of Noise Aware Audio Visual Speech Denoising,” describes a method for producing clearer speech from noisy audio-visual input. It models noise from audio data to create a reconstructed noise signal, then estimates that signal using audio and visual frames. The method divides the signal into windows, calculates each window’s energy, and applies a softmax operation to estimate noise strength and refine audio features. It then combines the refined audio features with visual features and passes the result through a decoder to generate denoised audio-visual speech. Parida is associated with 1 patent.
Career Highlights
Parida’s patent reflects work at the intersection of audio processing and visual information. By using visual frames alongside audio signals, the described approach aims to account for noise when reconstructing speech.
Collaborations
Parida’s listed coworkers include Gaurav Sharma and Siddharth Srivastava. His work at Tensortype Inc. provides the professional context for his patent activity.
Conclusion
Kranti Kumar Parida’s patent presents a noise-aware approach to audio-visual speech denoising, combining audio and visual features to produce clearer speech.
Data source: USPTO (United States Patent and Trademark Office) public patent records. Weekly synchronization. How IDiyas builds this profile