Audio device with uncertainty quantification and related methods
View Patent ↗An audio device comprising memory, an interface, and one or more processors, wherein the one or more processors are configured to obtain audio data; process the audio data for provision of an audio output; process the audio data for provision of one or more audio parameters indicative of one or more characteristics of the audio data; map the one or more audio parameters to a first latent space of a first neural network for provision of a mapping parameter indicative of whether the one or more audio parameters belong to a training manifold of the first latent space; determine, based on the mapping parameter, an uncertainty parameter indicative of an uncertainty of processing quality; and control the processing of the audio data for provision of the audio output based on the uncertainty parameter.
1. An audio device comprising memory, an interface, and one or more processors, wherein the one or more processors are configured to:
obtain audio data;
process the audio data for provision of an audio output;
process the audio data for provision of one or more audio parameters indicative of one or more characteristics of the audio data;
map the one or more audio parameters to a first latent space of a first neural network for provision of a mapping parameter indicative of whether the one or more audio parameters belong to a training manifold of the first latent space; determine, based on the mapping parameter, an uncertainty parameter indicative of an uncertainty of processing quality; and
control the processing of the audio data for provision of the audio output based on the uncertainty parameter.
2. The audio device according to claim 1 , wherein to process the audio data for provision of one or more audio parameters comprises to encode the audio data for provision of the one or more audio parameters using the first neural network.
3. The audio device according to claim 1 , wherein the one or more processors comprise a digital signal processor comprising a second neural network, and wherein to process the audio data for provision of an audio output comprises to provide the audio data as input to the second neural network and to process the audio data using the second neural network for provision of a primary output based on the uncertainty parameter, and wherein the audio output is based on the primary output.
4. The audio device according to claim 3 , wherein the second neural network comprises a deep neural network.
5. The audio device according to claim 3 , wherein the digital signal processor comprises a controller configured to determine a controller output based on the uncertainty parameter and wherein to control the processing of the audio data for provision of the audio output comprises to control the processing of the audio data based on the controller output.
6. The audio device according to claim 3 , wherein the one or more processors comprise a secondary processor different from the digital signal processor, wherein to process the audio data for provision of an audio output comprises to process the audio data for provision of a secondary output different from the primary output using the secondary processor, wherein the audio output is based on the secondary output.
7. The audio device according to claim 6 , wherein the one or more processors comprise a mixer, and wherein to process the audio data for provision of an audio output comprises to mix the primary output and the secondary output for provision of a mixed output, and wherein the audio output is based on the mixed output.
8. The audio device according to claim 1 , wherein the training manifold comprises a probability distribution, and wherein to map the one or more audio parameters comprises to map the one or more audio parameters to the probability distribution for provision of the mapping parameter.
9. The audio device according to claim 1 , wherein to map the one or more audio parameters with a first latent space of a first neural network comprises to determine a distance between the one or more audio parameters and the training manifold.
10. The audio device according to claim 1 , wherein the one or more processors are configured to:
determine whether the uncertainty parameter satisfies a first criterion, and in accordance with the uncertainty parameter satisfying the first criterion, process the audio data according to a first signal processing scheme for provision of the audio output.
11. The audio device according to claim 10 , wherein the one or more processors are configured to: in accordance with the uncertainty parameter not satisfying the first criterion and/or in accordance with the uncertainty parameter satisfying a second criterion, process the audio data according to a second signal processing scheme for provision of the audio output.
12. The audio device according to claim 11 , wherein the one or more processors are configured to: in accordance with the uncertainty parameter not satisfying the first criterion and in accordance with the uncertainty parameter not satisfying the second criterion, process the audio data according to a third signal processing scheme for provision of the audio output.
13. The audio device according to claim 1 , wherein the one or more processors are configured to output the audio output via the interface.
14. The method of operating an audio device, the method comprising:
obtaining audio data;
processing the audio data for provision of an audio output;
processing the audio data for provision of one or more audio parameters indicative of one or more characteristics of the audio data;
mapping the one or more audio parameters to a first latent space of a first neural network for provision of a mapping parameter indicative of whether the one or more audio parameters belong to a training manifold of the first latent space;
determining, based on the mapping parameter, an uncertainty parameter indicative of an uncertainty of processing quality; and
controlling the processing of the audio data for provision of the audio output based on the uncertainty parameter.
15. The method according to claim 14 , wherein processing the audio data for provision of one or more audio parameters comprises encoding the audio data for provision of the one or more audio parameters using the first neural network.
16. The method of claim 14 , wherein processing the audio data for provision of an audio output comprises providing the audio data as input to a second neural network and processing the audio data using the second neural network for provision of a primary output based on the uncertainty parameter, and wherein the audio output is based on the primary output.
17. The method of claim 16 , wherein the second neural network comprises a deep neural network.
18. The method of claim 16 , wherein controlling the processing of the audio data for provision of the audio output comprises controlling the processing of the audio data based on the controller output.
19. The method of claim 16 , wherein processing the audio data for provision of an audio output comprises processing the audio data for provision of a secondary output different from the primary output using the secondary processor, wherein the audio output is based on the secondary output.
20. The method of claim 16 ,
wherein processing the audio data for provision of an audio output comprises mixing the primary output and a secondary output for provision of a mixed output, and wherein the audio output is based on the mixed output.