Determination of composite acoustic parameter value for presentation of audio content

1. A method comprising:

determining a spatial acoustic parameter based in part on a signal-to-noise ratio (SNR) improvement value and a model that maps SNR improvement values to spatial acoustic parameters;

determining a temporal acoustic parameter based in part on audio signals from an acoustic sensor array; and

determining a composite acoustic parameter value based in part on the spatial acoustic parameter and the temporal acoustic parameter, wherein audio content presented to a user is adjusted based in part on the composite acoustic parameter value.

2. The method of claim 1 , wherein the composite acoustic parameter value is a value of an early-to late ratio or a direct-to-reverberant ratio.

3. The method of claim 1 , wherein at least one of the audio signals from an acoustic sensor of the acoustic sensor array includes a late reflection component.

4. The method of claim 1 , wherein the model is a machine learning model trained based on a plurality of SNR improvement values and a plurality of spatial acoustic parameters, each of the plurality of spatial acoustic parameters corresponding to one of the plurality of SNR improvement values.

5. The method of claim 1 , further comprising:

determining a value of a room acoustic parameter based on the composite acoustic parameter value,

wherein the audio content presented to the user is adjusted based in part on the room acoustic parameter.

6. The method of claim 1 , further comprising:

updating a virtual model of a local area based on the composite acoustic parameter value.

7. The method of claim 1 , wherein determining the composite acoustic parameter value based in part on the spatial acoustic parameter and the temporal acoustic parameter comprises:

determining a spatial weight for the spatial acoustic parameter;

determining a temporal weight for the temporal acoustic parameter; and

aggregating the spatial acoustic parameter and the temporal acoustic parameter based on the spatial weight and temporal weight.

8. The method of claim 7 , wherein at least one of the spatial weight and the temporal weight is determined based on a size of a local area, a number of acoustic sensors in the acoustic sensor array, or a combination of both.

9. A method comprising:

determining a spatial acoustic parameter based in part on a signal-to-noise ratio (SNR) improvement value and a model that maps SNR improvement values to spatial acoustic parameters;

determining a temporal acoustic parameter based in part on audio signals from an acoustic sensor array;

determining a composite acoustic parameter value based in part on the spatial acoustic parameter and the temporal acoustic parameter;

adjusting audio content based in part on the composite acoustic parameter value; and

presenting the adjusted audio content to a user.

10. The method of claim 9 , wherein adjusting the audio content based in part on the composite acoustic parameter value comprises:

adjusting the audio content based in part on the composite acoustic parameter value such that the audio content appears to the user as originating from a sound source in a local area.

11. The method of claim 9 , wherein the composite acoustic parameter value is a value of an early-to late ratio or a direct-to-reverberant ratio.

12. The method of claim 9 , further comprising:

determine a value of a room acoustic parameter based on the composite acoustic parameter value,

wherein adjusting the audio content based in part on the composite acoustic parameter value comprises adjusting the audio content based on the composite acoustic parameter value and the value of the room acoustic parameter.

13. The method of claim 9 , wherein determining the composite acoustic parameter value based in part on the spatial acoustic parameter and the temporal acoustic parameter comprises:

determining a spatial weight for the spatial acoustic parameter;

determining a temporal weight for the temporal acoustic parameter; and

aggregating the spatial acoustic parameter and the temporal acoustic parameter based on the spatial weight and temporal weight.

14. The method of claim 13 , wherein at least one of the spatial weight and the temporal weight is determined based on a size of a local area, a number of acoustic sensors in the acoustic sensor array, confidence values of the spatial acoustic parameter and the temporal acoustic parameter, a signal-to-noise ratio of a directionally enhanced audio signal, the SNR improvement value, a white-noise gain, a number of sound sources in the local area, types of noises in the local area, a type of the local area, a speed of movement of the acoustic sensor array, an orientation of the acoustic sensor array, one or more qualities of acoustic sensors in the acoustic sensor array, a temporal sampling rate, spatial position of acoustic sensors on the acoustic sensor array, a spatial aliasing frequency, or some combination thereof.

15. A non-transitory computer readable storage medium configured to store program code instructions, when executed by a processor of a device, cause the device to:

determine a spatial acoustic parameter based in part on a signal-to-noise ratio (SNR) improvement value and a model that maps SNR improvement values to spatial acoustic parameters;

determine a temporal acoustic parameter based in part on audio signals from an acoustic sensor array; and

determine a composite acoustic parameter value based in part on the spatial acoustic parameter and the temporal acoustic parameter, wherein audio content presented to a user is adjusted based in part on the composite acoustic parameter value.

16. The computer-readable storage medium of claim 15 , wherein the composite acoustic parameter value is a value of an early-to late ratio or a direct-to-reverberant ratio.

17. The computer-readable storage medium of claim 15 , wherein at least one of the audio signals from an acoustic sensor of the acoustic sensor array includes a late reflection component.

18. The computer-readable storage medium of claim 15 , wherein the model is a machine learning model trained based on a plurality of SNR improvement values and a plurality of spatial acoustic parameters, each of the plurality of spatial acoustic parameters corresponding to one of the plurality of SNR improvement values.

19. The computer-readable storage medium of claim 15 , further comprising stored instructions that when executed cause the device to:

determine a value of a room acoustic parameter based on the composite acoustic parameter value,

wherein the audio content presented to the user is adjusted based in part on the room acoustic parameter.

20. The computer-readable storage medium of claim 15 , further comprising stored instructions that when executed further cause the device to:

determine a spatial weight for the spatial acoustic parameter;

determine a temporal weight for the temporal acoustic parameter; and

aggregate the spatial acoustic parameter and the temporal acoustic parameter based on the spatial weight and temporal weight.