Hierarchical environmental classification in a hearing prosthesis
Presented herein are techniques for generating a hierarchical classification of a set of sound signals received at hearing prosthesis. The hierarchical classification includes a plurality of nested classifications of a sound environment associated with the set of sound signals received at hearing prosthesis, including a primary classification and one or more secondary classifications that each represent different characteristics of the sound environment. The primary classification represents a basic categorization of the sound environment, while the secondary classifications define sub-categories/refinements of the associated primary classification and/or other secondary classifications.
1. A method, comprising:
receiving a set of input signals;
determining, with a first classifier, a first classification of the set of input signals;
conditionally activating a second classifier based on the first classification; and
determining, using the second classifier, at least one second classification of the set of input signals, wherein the at least one second classification is a sub-category of the first classification.
2. The method of claim 1 , further comprising:
generating a classification output representing the first classification and the at least one second classification; and
storing the classification output as part of a classification data set.
3. The method of claim 1 , further comprising:
adjusting one or more settings of a device based on the first classification and the at least one second classification.
4. The method of claim 3 , wherein the device is a hearing device, and wherein the one or more settings comprise at least one of: noise reduction settings, tinnitus masking settings, microphone settings, gain settings, channel dynamic range, maxima selection, and comfort settings.
5. The method of claim 1 , wherein the first classification is a classification of an ambient environment.
6. The method of claim 5 , wherein the first classification indicates that the ambient environment is an environment that includes speech, and wherein determining the at least one second classification comprises:
sub-categorizing the ambient environment based on attributes of the speech.
7. The method of claim 5 , wherein the first classification indicates that the ambient environment is an environment that includes noise, and wherein determining the at least one second classification comprises:
sub-categorizing a type of noise in the ambient environment.
8. The method of claim 1 , further comprising:
generating, based on the set of input signals, stimulation signals for delivery to a recipient of a medical device.
9. A device, comprising:
one or more input elements configured to receive at least one set of environmental signals;
a memory; and
one or more processors coupled to the memory and the one or more input elements, and configured to:
convert the at least one set of environmental signals into one or more output signals for use in delivering stimulation to a recipient of the device;
determine a primary classification of the at least one set of environmental signals as a first classification; and
in response to determining the primary classification of the at least one set of environmental signals as the first classification, determine a secondary classification of the at least one set of environmental signals as a second classification to generate a complex classification comprising a plurality of nested classifications of an ambient environment associated with the at least one set of environmental signals.
10. The device of claim 9 , wherein the one or more processors are configured to determine an additional secondary classification of the at least one set of environmental signals as a third classification in response to determining the secondary classification of the at least one set of environmental signals is the second classification, wherein the third classification sub-categorizes the second classification.
11. The device of claim 9 , wherein the first classification indicates that the ambient environment is an environment that includes speech, and wherein to determine the secondary classification, the one or more processors are configured to:
sub-categorize the ambient environment based on attributes of the speech.
12. The device of claim 9 , wherein the first classification indicates that the ambient environment is an environment that includes noise, and wherein to determine the secondary classification, the one or more processors are configured to:
sub-categorize a type of noise in the ambient environment.
13. The device of claim 9 , wherein the one or more processors are configured to:
generate a classification output representing the plurality of nested classifications; and
store the classification output as part of a classification data set in the memory.
14. The device of claim 9 , wherein the one or more processors are configured to:
automatically adjust one or more settings of the device based on the plurality of nested classifications.
15. The device of claim 14 , wherein the one or more settings comprise at least one of:
noise reduction settings, tinnitus masking settings, microphone settings, gain settings, channel dynamic range, maxima selection, and comfort settings.
16. The device of claim 9 , wherein the one or more processors are configured to:
block determination of the secondary classification of the at least one set of environmental signals in response to determining the primary classification of the at least one set of environmental signals is a third classification.
17. One or more non-transitory computer readable storage media comprising instructions that, when executed by a processor, cause the processor to:
obtain a plurality of input signals;
perform a first classification of the plurality of input signals; and
conditionally perform at least one second classification of the plurality of input signals based on the first classification, wherein the at least one second classification is a sub-category of the first classification.
18. The one or more non-transitory computer readable storage media of claim 17 , further comprising instructions operable to:
generate a classification output representing the first classification and the at least one second classification; and
store the classification output as part of a classification data set in memory.
19. The one or more non-transitory computer readable storage media of claim 17 , further comprising instructions operable to:
adjust one or more settings of a device based on the first classification and the at least one second classification.
20. The one or more non-transitory computer readable storage media of claim 17 , further comprising instructions operable to:
generate at least one output signal based on the plurality of input signals.