Effects of Noise Reduction on AM and FM Perception

Abstract

The goal of noise reduction (NR) algorithms in digital hearing aid devices is to reduce background noise whilst preserving as much of the original signal as possible. These algorithms may increase the signal-tonoise ratio (SNR) in an ideal case, but they generally fail to improve speech intelligibility. However, due to the complex nature of speech, it is difficult to disentangle the numerous low-and high-level effects of NR that may underlie the lack of speech perception benefits. The goal of this study was to better understand why NR algorithms do not improve speech intelligibility by investigating the effects of NR on the ability to discriminate two basic acoustic features, namely amplitude modulation (AM) and frequency modulation (FM) cues, known to be crucial for speech identification in quiet and in noise. Here, discrimination of complex, non-linguistic AM and FM patterns was measured for normal hearing listeners using a same/different task. The stimuli were generated by modulating 1-kHz pure tones by either a two-component AM or FM modulator with patterns changed by manipulating component phases. Modulation rates were centered on 3 Hz. Discrimination of AM and FM patterns was measured in quiet and in the presence of a white noise that had been passed through a gammatone filter centered on 1 kHz. The noise was presented at SNRs ranging from −6 to +12 dB. Stimuli were left as such or processed via an NR algorithm based on the spectral subtraction method. NR was found to yield small but systematic improvements in discrimination for the AM conditions at favorable SNRs but had little effect, if any, on FM discrimination. A computational model of early auditory processing was developed to quantify the fidelity of AM and FM transmission. The model captured the improvement in discrimination performance for AM stimuli at high SNRs with NR. However, the model also predicted a relatively small detrimental effect of NR for FM stimuli in contrast with the average psychophysical data. Overall, these results suggest that the lack of benefits of NR on speech intelligibility is partly caused by the limited effect of NR on the transmission of narrowband speech modulation cues.

Effects of Noise Reduction on AM Perception for Hearing-Impaired Listeners

“…The stimuli and task were similar to that described in Ives et al (2013). Listeners were required to discriminate pairs of stimuli that differed in their AM patterns.…”

“…While the nature of the benefit of such systems is still debated (e.g., lowering noise annoyance, reducing cognitive load or improving ease of listening; Brons et al 2013;Palmer et al 2006;Sarampalis et al 2009), it is clear that NR does not improve speech intelligibility compared to the unprocessed conditions (Hu and Loizou 2007). Ives et al (2013) investigated the effectiveness of a modern NR algorithm Malah 1984, 1985) on the transmission of slow amplitude-modulation (AM) and frequency-modulation (FM) cues; these are known to be important for speech recognition in quiet and in noise (e.g. Houtgast and Steeneken 1985;Shannon et al 1995;Zeng et al 2005).…”

“…The current study aimed to assess whether the small, but significant beneficial effects of NR on the discrimination of AM cues observed for NH listeners could also be observed for hearing-impaired (HI) listeners. The discrimination of FM cues for HI listeners was not assessed as Ives et al (2013) had shown the benefit from NR on FM cues was weak. Throughout this study, a single-microphone spectral-subtraction NR algorithm was used (see section II.…”

“…Cochlear damage can also be associated with increased (internal) variability in the encoding process which is usually modeled by increased internal (neural) noise (Zwislocki and Jordan 1986;Derleth et al 2001). In the current study, the computational model of early auditory processing developed by Ives et al (2013) to quantify the effects of NR on the fidelity of AM transmission was further developed and used to determine the relative contribution of these acoustical and sensory factors.…”

Perception of frequency modulation is mediated by cochlear place coding

“…The perception of FM is often degraded in older listeners and people with hearing loss Fougère and Demany, 1998 Thomson et al, 2017). Current assistive listening devices, such as 50 hearing aids and cochlear implants, have been generally unsuccessful at reintroducing viable FM 51 cues to the auditory system (Chen and Zeng, 2004;Ives et al, 2013). This lack of success is 52 partly related to a gap in our scientific understanding regarding how FM is extracted by the brain 53 from the information available in the auditory periphery.…”