Alternative Control Technologies: Human Factors Issues

Active soundproof earmuffs system with passive noise insulation structure compensation

Localization algorithms have become of considerable interest for robot audition, acoustic navigation, teleconferencing, speaker localization, and many other applications over the last decade. In this paper, we present a real-time implementation of a Gaussian mixture model (GMM) based probabilistic sound source localization algorithm for a low-power VLIW-SIMD processor for hearing devices. The algorithm has been proven to allow for robust localization of multiple sound sources simultaneously in reverberant and noisy environments. Real-time computation for audio frames of 512 samples at 16 kHz was achieved by introducing algorithmic optimizations and hardware customizations. To the best of our knowledge, this is the first real-time capable implementation of a computationally complex GMM-based sound source localization algorithm on a low-power processor. The resulting estimated core area without consideration of memory in 40nm low-power TSMC technology is 188,511 pm2.

Real-time implementation of a GMM-based binaural localization algorithm on a VLIW-SIMD processor

Griffiths’ Variable Learning Rate Online Sequential Learning Algorithm for Feed-Forward Neural Networks

Development of Voice Activated Ground Control Station

Robust Active Noise Control System for Fighter Aircraft Pilot Helmet Application

In this paper, we have chronicled the development of sound localization system based on TDOA (Time Difference of Arrival). Acoustic Source Localization (ASL) is a technique used to track and locate the exact location of a sound source using an array of microphones. The concept of ASL uses sound signals captured from an array of microphones and they are processed using TDOA localization method to estimate the probable direction of sound source w.r.t to the microphone location. TDOA algorithm is a time delay estimation technique which estimates the time difference in the signal received at each microphone pair. These time delays obtained are then used in the Linear Least Squares (LSQR) Algorithm to estimate the source position w.r.t the microphone array. This system is implemented in real-time by using an on-board DSP processor TMS320C6748.

Real time acoustic source localization of emergency signals

The main idea of this paper is to bring out the uniqueness of the speech corpus required for development of command and control applications such as voice-controlled MAV. Since MAV finds application in adverse environments, the effect of noise degrades ASR performance. Since English words uttered are greatly influenced by user’s mother tongue, there is a necessity to create a customized speech corpus. The corpus creation is accomplished by a NALVoiceCorpus tool, which is designed to capture the specific requirements of the corpus. The tool is quite generic in nature and it can find application in development of any ASR system.

H. Lokesha

Papers

Development of Voice Activated Ground Control Station

Robust Active Noise Control System for Fighter Aircraft Pilot Helmet Application

Real time acoustic source localization of emergency signals

Speech Corpus Development for Voice-Controlled MAV