Privacy protection in collective feedforward
Personal audio systems and methods are disclosed. A personal audio system includes a processor to generate a personal audio stream by processing an ambient audio stream in accordance with an active processing parameter set, a circular buffer memory to store a most recent snippet of the ambient audio stream, and an event detector to detect a trigger event. In response to detection of the trigger event, a controller may extract audio feature data from the most recent snippet of the ambient audio stream and transmit the audio feature data and associated metadata to a knowledgebase remote from the personal audio system.
1. A system, comprising:
a memory; and
a processor configured to:
analyze audio feature data and associated metadata received from a personal audio system, wherein the audio feature data is extracted from a most recent snippet of an ambient audio stream;
develop a current ambient sound profile based, at least in part, on the audio feature data, the current ambient sound profile including values for one or more features selected from a group of features consisting of: an overall loudness level; a loudness of one or more frequency bands; a spectral envelope shape; one or more spectrographic features; an indicator of the presence or absence of odd harmonics, even harmonics, or both odd and even harmonics; a normalized or absolute loudness of noise; and a low-frequency periodicity;
develop a new processing parameter set based, at least in part, on the current ambient sound profile, the new processing parameter set indicating a type and degree of one or more processes to be performed on the ambient audio stream;
provide the new processing parameter set to the personal audio system; and
cause the memory to store the new processing parameter set and the current ambient sound profile.
2. The system of claim 1 , wherein the processor is further configured to receive the audio feature data in response to a trigger event associated with the personal audio system.
3. The system of claim 2 , wherein the trigger event includes retrieving a selected processing parameter set from a dataset memory of the personal audio system for use as an active processing parameter set.
4. The system of claim 2 , wherein
the trigger event includes receiving, via a user interface, a command to store an active processing parameter set in a dataset memory, and
the associated metadata transmitted to the device includes the active processing parameter set.
5. The system of claim 1 , wherein the associated metadata includes the active processing parameter set before modification.
6. The system of claim 1 , wherein the associated metadata transmitted to the device includes a descriptor received from a user via a user interface.
7. The system of claim 1 , wherein the associated metadata transmitted to the device includes data indicating the location of the personal audio system.
8. The system of claim 1 , wherein the audio feature data includes one or more of arithmetic mean, geometric mean, harmonic mean, centroid, variance, standard deviation, spectral skew, kurtosis, spectral shape, rolloff, spread, flatness, spectral flux, Mel frequency cepstral coefficients, pitch, tonal power ratio, harmonic-to-average power ratio, maximum of autocorrelation function, zero crossing rate, RMS power, peak power, crest factor, amplitude/power envelope and attack/decay rates.
9. The system of claim 1 , wherein the processor is configured to develop the new processing parameter set using one or more machine learning techniques.
10. The system of claim 1 , wherein the processor is configured to develop the new processing parameter set using one or more analytical techniques.
11. The system of claim 1 , wherein to develop a new processing parameter set includes:
sorting the audio feature data and associated metadata from the personal audio system and additional audio feature data and associated metadata from one or more other personal audio systems into a plurality of clusters; and
developing a corresponding consensus processing parameter set for each cluster of the plurality of clusters.
12. The system of claim 11 , wherein the plurality of clusters are sorted based on the associated metadata from the personal audio system and the associated metadata from the other personal audio systems.
13. The system of claim 11 , wherein the corresponding consensus processor parameter set is based on a median or average of processing parameters associated with a cluster.
14. The system of claim 11 , wherein an outlier processing parameter is discarded from a cluster.
15. A method, comprising:
receiving one or more user commands to modify an active processing parameter set for a personal audio system or to create a new processing parameter set for the personal audio system;
receiving audio feature data and associated metadata received from the personal audio system, wherein the audio feature data is extracted from a most recent snippet of an ambient audio stream;
analyzing the audio feature data and the associated metadata;
developing a new processing parameter set based on the audio feature data, the new processing parameter set indicating a type and degree of one or more processes to be performed on the ambient audio stream; and
providing the new processing parameter set to the personal audio system and one or more other personal audio systems.
16. The method of claim 15 , further comprising receiving the audio feature data in response to a trigger event associated with the personal audio system.
17. The method of claim 15 , wherein developing a new processing parameter set includes:
sorting the audio feature data and associated metadata from the personal audio system and additional audio feature data and associated metadata from one or more other personal audio systems into a plurality of clusters; and
developing a corresponding consensus processing parameter set for each cluster of the plurality of clusters.
18. The method of claim 17 , wherein the plurality of clusters are sorted based on the associated metadata from the personal audio system and the associated metadata from the other personal audio systems.
19. The method of claim 17 , wherein the corresponding consensus processor parameter set is based on a median or average of processing parameters associated with a cluster.
20. A computer program product, the computer program product being embodied in a tangible non-transitory computer readable storage medium and comprising computer instructions for:
receiving one or more user commands to modify an active processing parameter set for a personal audio system or to create a new processing parameter set for the personal audio system;
receiving audio feature data and associated metadata received from the personal audio system, wherein the audio feature data is extracted from a most recent snippet of an ambient audio stream;
analyzing the audio feature data and the associated metadata;
developing a new processing parameter set based on the audio feature data, the new processing parameter set indicating a type and degree of one or more processes to be performed on the ambient audio stream; and
providing the new processing parameter set to the personal audio system.