Environment dependent noise tracking for speech enhancement

Abstract

Numerous efforts have focused on the problem of reducing the impact of noise on the performance of various speech systems such as speech recognition, speaker recognition, and speech coding. These approaches consider alternative speech features, improved speech modeling, or alternative training for acoustic speech models. This study presents an alternative viewpoint by approaching the same problem from the noise perspective. Here, a framework is developed to analyze and use the noise information available for improving performance of speech systems. The proposed framework focuses on explicitly modeling the noise and its impact on speech system performance in the context of speech enhancement. The framework is then employed for development of a novel noise tracking algorithm for achieving better speech enhancement under highly evolving noise types. The first part of this study employs a noise update rate in conjunction with a target enhancement algorithm to evaluate the need for tracking in many enhancement algorithms.

A Hybrid Coherence Model for Noise Reduction in Reverberant Environments

“…The complex coherence function between the two input signals is defined as (2) where denotes the cross-power spectral density (CSD), and the power spectral density (PSD). In the above equation, the index was omitted for sake of clarity.…”

“…2 shows a comparison between the (true) coherence function of the noisy signals computed by (2), and the predicted (approximation) obtained using the models given in (7) and (8). This comparison was done with simulation of signals at two microphones with 18 mm distance, at sampling rate of 16 kHz.…”

“…The concept of "Environmental Sniffing" has recently emerged as an important domain to improve the robustness of speech and language processing technology. The hearing aid and cochlear implant (CI) research community has long recognized the importance of environmental classification [1], [2]. In general, two types of noise fields are common in microphone array speech processing studies, which include coherent and diffuse noise fields [3].…”

Robust and efficient environment detection for adaptive speech enhancement in cochlear implants

“…In addition, knowing the type of noise/distortion can be employed to adapt the necessary noise frame update rate needed to achieve a required level performance [17].…”