Sound processing based on a confidence measure
AssigneeCochlear Limited
InventorsAdam A. Hersbach, Stefen J. Mauger, John M. Heasman, Pam W. Dawson
Life of this patent
Estimating the term…
Abstract
A method for processing sound that includes, generating one or more noise component estimates relating to an electrical representation of the sound and generating an associated confidence measure for the one or more noise component estimates. The method further comprises processing, based on the confidence measure, the sound.
Claims22
A sound processing method, comprising:
- generating an electrical representation of a received sound having desired signal components and noise components;
- generating at least one estimate of a noise component;
- generating a confidence measure of the noise component estimate;
- and
- processing the electrical representation at least partially based on the confidence measure.
A sound processing method including:
- receiving a plurality of input signals representing a plurality of spectral components of a sound;
- selecting one or more of the input signals by processing a value representing the energy of each spectral component and a value indicating a psychoacoustic importance of each spectral component;
- and
- processing the one or selected input signals to generate stimuli for delivery to a recipient via an electrode array of a hearing prosthesis.
A method of processing an electrical representation of a sound so as to generate electrical stimulation signals for delivery to a recipient via electrodes in an implantable electrode array, comprising:
- generating a signal-to-noise ratio (SNR) estimate for at least a component of the electrical representation;
- processing the SNR estimate for said component using a function that varies with the component's estimated SNR to determine a gain level corresponding to said component, wherein for an estimated instantaneous SNR of between −5 dB and 20 dB at least a portion of the gain function lies in a region bounded by a gain function defined by
- Gw
-
- (
- ξ
- )
- =
- (
- ξ
-
- (
- t
- ,
- f
- )
- ξ
-
- (
- t
- ,
- f
- )
- +
- 0.12
- )
- 20
- and a gain function defined by
- Gw
-
- (
- ξ
- )
- =
- (
- ξ
-
- (
- t
- ,
- f
- )
- ξ
-
- (
- t
- ,
- f
- )
- +
- 1
- )
- 20
- where Gw is the gain level, ζ is the SNR estimate.
Dependencies are inferred from the claim text.