Search PubMed⌕ Search
Home/Archive/PubMed

PubMed · 15237827

An efficient robust sound classification algorithm for hearing aids.

Peter Nordqvist·Arne Leijon
View original source ↗Export metadata ↓

Abstract

An efficient robust sound classification algorithm based on hidden Markov models is presented. The system would enable a hearing aid to automatically change its behavior for differing listening environments according to the user's preferences. This work attempts to distinguish between three listening environment categories: speech in traffic noise, speech in babble, and clean speech, regardless of the signal-to-noise ratio. The classifier uses only the modulation characteristics of the signal. The classifier ignores the absolute sound pressure level and the absolute spectrum shape, resulting in an algorithm that is robust against irrelevant acoustic variations. The measured classification hit rate was 96.7%-99.5% when the classifier was tested with sounds representing one of the three environment categories included in the classifier. False-alarm rates were 0.2%-1.7% in these tests. The algorithm is robust and efficient and consumes a small amount of instructions and memory. It is fully possible to implement the classifier in a DSP-based hearing instrument.

Explore related subjects

Acoustic StimulationAlgorithmsEnvironmentFemaleHearing AidsHumansMaleMarkov ChainsNoisePersons with Hearing DisabilitiesSignal Processing, Computer-AssistedSpeech PerceptionJournal ArticleResearch Support, Non-U.S. Gov't

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Peter Nordqvist, Arne Leijon. 2004. An efficient robust sound classification algorithm for hearing aids.. https://doi.org/10.1121/1.1710877

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related citations

PubMed2005

A flexible auditory research platform using acoustic or electric stimuli for adults and young children.

A user-friendly and versatile research platform for use in auditory experiments, referred to as APEX (Application for PsychoElectrical eXperiments), is described. The platform takes care of automatic stimulus presentation and collection of the subject's responses. Acoustical auditory, as well as electrical auditory experiments with CI recipients can be conducted. The platform currently supports LAURA, Nucleus CI22 and Nucleus CI24 cochlear implants. The graphical user interface for the subjects has been extended to allow for testing very young children, by embedding the psychophysical procedures in a computer game. The research platform is available free of charge.

Acoustic Stimulation↗
PubMed2005

Haloperidol can increase responding to both discrete and contextual cues in trace conditioned rats.

Haloperidol has been shown to enhance attentional selectivity in conditioning procedures. For example, in latent inhibition (LI) it improves animals' ability to treat as irrelevant, stimuli that have previously been presented without consequence. The present study tested whether this finding would generalize to other procedures that present animals with weak predictors. We therefore used a trace conditioning procedure to present rats with a conditioned stimulus (CS) weakened through temporal discontiguity (rather than preexposure in LI) and a flashing light background provided an alternative experimental stimulus. In Experiment 1, a noise CS was paired contiguously (at '0 s') with food or at a 10 s trace interval. In Experiment 2, the trace interval was lengthened to 20 s. In both experiments, haloperidol treatment generally reduced responding in 0 s contiguous groups. By contrast, 0.03 mg/kg haloperidol enhanced conditioning, selectively, to the weakly predictive trace CS, though it was without effect on responding within the trace interval. In addition, again at 0.03 mg/kg, haloperidol significantly increased excitatory conditioning to contextual stimuli in trace groups relative to contiguous groups. At the shorter (10 s) Experiment 1 trace, this result was shown in the extinction test of conditioning to the background stimulus. At the longer (20 s) Experiment 2 trace, this result was shown in the acquisition of responding to the box context in the inter-trial-interval. The demonstration that low dose haloperidol can increase conditioning is novel. This increase was seen selectively with stimuli (both trace-conditioned and contextual) that should have been treated as weak predictors so these results are contrary to what was expected on the basis of haloperidol effects on stimuli weakened through pre-exposure. The possibility that increased contextual conditioning could be relevant to the interpretation of haloperidol-induced enhancement of LI is discounted. However, it is suggested that this result could nonetheless reflect cognitive enhancement.

Acoustic Stimulation↗
PubMed2005

Association between violent behaviour and impaired prepulse inhibition of the startle response in antisocial personality disorder and schizophrenia.

Violent behaviour has a strong association with antisocial personality disorder (APD) and schizophrenia. Although developments in the understanding of socio-environmental factors associated with violence should not be ignored, advances in prevention and treatment of violent behaviour would benefit by improved understanding of its neurobiological and cognitive basis. The authors, therefore, investigated prepulse inhibition (PPI) of the startle response in APD and schizophrenia in relation to a history of serious violence. The neural substrates of PPI, especially the hippocampus, amygdala, thalamus and basal ganglia, are implicated in violence as well as in APD and schizophrenia. The study included four groups: (i) patients with APD and a history of violence, (ii) patients with schizophrenia and a history of violence, (iii) patients with schizophrenia without a history of violence, and (iv) healthy subjects with no history of violence or a mental disorder. All subjects were assessed identically on acoustic PPI. Compared to healthy subjects, significantly reduced PPI occurred in APD, violent schizophrenia and non-violent schizophrenia patients. Although PPI did not significantly differentiate the three clinical groups, high ratings of violence were modestly associated with reduced PPI across the entire study sample. Violent patients with impulsive and premeditated violence showed comparable PPI. The association between violent behaviour and impaired PPI suggests that neural structures and functions underlying PPI are implicated in (inhibition of) violence.

Acoustic Stimulation↗