Multi-Speaker Localization Using Convolutional Neural Network Trained with Noise

12/12/2017
by   Soumitro Chakrabarty, et al.
0

The problem of multi-speaker localization is formulated as a multi-class multi-label classification problem, which is solved using a convolutional neural network (CNN) based source localization method. Utilizing the common assumption of disjoint speaker activities, we propose a novel method to train the CNN using synthesized noise signals. The proposed localization method is evaluated for two speakers and compared to a well-known steered response power method.

READ FULL TEXT
research
07/31/2018

Multi-Speaker DOA Estimation Using Deep Convolutional Networks Trained with Noise Signals

Supervised learning based methods for source localization, being data dr...
research
05/26/2017

Text-Independent Speaker Verification Using 3D Convolutional Neural Networks

In this paper, a novel method using 3D Convolutional Neural Network (3D-...
research
05/02/2017

Broadband DOA estimation using Convolutional neural networks trained with noise signals

A convolution neural network (CNN) based classification method for broad...
research
12/18/2019

The CNN-based Coronary Occlusion Site Localization with Effective Preprocessing Method

The Coronary Artery Occlusion (CAO) acutely comes to human, and it highl...
research
11/04/2021

Extended Abstract Version: CNN-based Human Detection System for UAVs in Search and Rescue

This paper proposes an approach for the task of searching and detecting ...
research
06/02/2017

Prosodic Event Recognition using Convolutional Neural Networks with Context Information

This paper demonstrates the potential of convolutional neural networks (...
research
03/20/2020

A deep learning approach for lower back-pain risk prediction during manual lifting

Occupationally-induced back pain is a leading cause of reduced productiv...

Please sign up or login with your details

Forgot password? Click here to reset