A Dual-Microphone Algorithm That Can Cope With Competing-Talker Scenarios

Abstract

Abstract-This paper introduces a novel technique for signal-to-noise ratio (SNR) estimation for scenarios where two closely-spaced microphones are available. The proposed technique utilizes the real and imaginary parts of the coherence function between the input signals to estimate the SNR without assuming prior knowledge of the noise statistics. The corresponding dual-microphone speech enhancement algorithm utilizes a Wiener filter as a gain function constructed using the SNR values computed by the coherence function. Since the proposed SNR estimation technique does not require access to noise statistics, it can be applied in situations where interfering speakers are present. An adaptive speech reception threshold (SRT) test was used to assess the intelligibility of speech processed by the proposed algorithm in scenarios where one or two interfering talkers were present in anechoic and reverberant conditions. Intelligibility listening tests were conducted with both normal-hearing (NH) and cochlear implant (CI) listeners. Results revealed significant improvements in intelligibility and quality over a (baseline) fixed directional algorithm and a well-established beamformer algorithm. In a nearly anechoic room with competing talkers, the improvement in SRT obtained relative to the directional microphone ranged from 5-10 dB, while the improvement obtained by the beamformer was about 2 dB. In reverberant environments, the improvement in SRT remained high (4-7 dB) at ms, and decreased to 1-2 dB at ms. Overall, the proposed algorithm provided significant benefits in intelligibility in anechoic and mildly reverberant environments making it suitable for hearing aid and cochlear implant applications.

A Hybrid Coherence Model for Noise Reduction in Reverberant Environments

“…In [8], we proposed a coherence-based dual-microphone noise reduction technique and showed that in anechoic (also low reverberant) rooms, where the noise field is highly coherent, it offers significant improvements over a fixed directional microphone and a well-established beamformer in terms of intelligibility and quality. We also observed that algorithm performance starts to degrade when tested in more reverberant rooms.…”

Two-microphone spatial filtering provides speech reception benefits for cochlear implant users in difficult acoustic environments

“…The 2.2 to 7.0 dB SRT benefit of Fennec over Dipole processing in a moderately-reverberant environment (T 60 ¼ 350 ms) compares favorably to the SRT benefit for the related processing technique of Yousefian and Loizou (2013), which uses estimated input coherence to estimate the input SNR which is then processed through a Wiener filter to yield the particular time/ frequency attenuation terms [analogous to those generated by Eq. (3) for Fennec].…”

“…This approach is based on direct analysis of phase differences, rather than the use of coherence as suggested by Yousefian and Loizou (2013). This distinction can yield performance differences, particularly in more complex environments where the target speech is degraded by reverberation and by multiple, simultaneous, noise sources.…”

“…This distinction can yield performance differences, particularly in more complex environments where the target speech is degraded by reverberation and by multiple, simultaneous, noise sources. Hersbach et al (2013) suggested it was this dependence on coherence which caused the Yousefian and Loizou (2013) algorithm to deteriorate in reverberant environments. Specifically, speech reception benefits provided by the coherence-based algorithm decreased from 5 to 10 dB in an anechoic environment to 0 to 2 dB in a moderately reverberant room (T 60 ¼ 465 ms).…”

A beamformer post-filter for cochlear implant noise reduction

“…These assumptions, however, generally do not hold for many real situations and, therefore, these theoretical approaches provide limited practical solutions to improve beamformer performance in cochlear implant devices. Yousefian and Loizou (2013) proposed a dual-microphone coherence-based noise reduction algorithm for cochlear implant users and demonstrated 5-10 dB SRT improvement over a fixed directional microphone in an anechoic room with one or two competing talkers. The algorithm assumed that speech and noise were coherent across the microphone array, and was also evaluated by normal hearing listeners in various degrees of reverberation.…”