Browsing by Author "Ganguly, Anshuman"
Now showing 1 - 2 of 2
- Results Per Page
- Sort Options
Item Noise-Robust Speech Source Localization and Tracking Using Microphone Arrays for Smartphone-Assisted Hearing Aid Devices(2018-05) Ganguly, Anshuman; 0000-0003-4409-3088 (Ganguly, A); Panahi, Issa M. S.Speech Source Localization (SSL) (or Direction of Arrival estimation) is a powerful pre-processing tool that helps identify the direction of the talker of interest in a noisy environment using multiple fixed microphones (known as a Microphone Array). This information is very helpful to the speech-processing pipeline and can be utilized to improve the performance of the overall system. With recent advancements, smartphones now possess the requisite hardware and computational power to perform real-time SSL for different applications. In this work, we propose application-specific SSL algorithms for three types of microphone arrays and show their effectiveness for smartphone implementation under realistic background noise conditions. We evaluate our proposed approaches in several realistic noisy conditions and present object evaluations to demonstrate the effectiveness of the proposed methods. We also propose the real-time implementation of some of our methods on the latest smartphones and smartphone-assisted devices.Item Real-Time Smartphone Implementation of Noise-Robust Speech Source Localization Algorithm for Hearing Aid Users(Acoustical Society of America) Ganguly, Anshuman; Kucuk, Abdullah; Panahi, Issa; Ganguly, Anshuman; Kucuk, Abdullah; Panahi, IssaSpeech source localization has numerous application areas such as hearing aid devices (HAD) and consumer electronics applications. Utilizing the powerful processing hardware of smartphones, we demonstrate that smartphones are capable of instantaneous estimation of sound location. In this paper, we present instantaneous direction of arrival (DOA) by using traditional Generalized Cross Correlation (GCC) followed by a spatial post-filtering stage. A simple voice activity detector (VAD) is used for the post-filtering stage to improve noise robustness in some realistic reverberant noisy environments. Root mean square error (RMSE) is used as an evaluation criterion for the proposed method. Both real recorded data and simulated data under different noise types are used for experiments. A real-time implementation of the method on an Android-based smartphone is also presented.