Detecting and handling driving event sounds during a navigation session

Abstract

To identify driving event sounds during navigation, a client device in a vehicle provides a set of navigation directions for traversing from a starting location to a destination location along a route. During navigation to the destination location, the client device identifies audio that includes a driving event sound from within the vehicle or an area surrounding the vehicle. In response to determining that the audio includes the driving event sound, the client device determines whether the driving event sound is artificial. In response to determining that the driving event sound is artificial, the client device presents a notification to the driver indicating that the driving event sound is artificial or masks the driving event sound to prevent the driver from hearing the driving event sound.

What is claimed is: 
     
         1 . A method for training a machine learning model to determine whether a driving event sound is real or artificial, the method comprising:
 obtaining, by one or more processors, a set of driving event sound characteristics for each of a plurality of driving event sounds;   for each driving event sound in the plurality of driving event sounds, obtaining, by the one or more processors, an indication of whether the driving event sound is from a real or artificial source; and   training, by the one or more processors, a machine learning model to determine whether a driving event sound is real or artificial using (i) the set of driving event sound characteristics corresponding to each driving event sound, and (ii) the indication of whether each driving event sound is from the real or artificial source,   wherein each set of driving event sound characteristics is classified according to whether the set corresponds to one of the plurality of driving event sounds from the real source or from the artificial source.   
     
     
         2 . The method of  claim 1 , further comprising:
 obtaining, by the one or more processors, audio playback data from an application executing on a client device;   obtaining, by the one or more processors, audio playback data from a device communicatively coupled to the client device; or   obtaining, by the one or more processors, ambient audio.   
     
     
         3 . The method of  claim 2 , further comprising:
 applying, by the one or more processors, the audio playback data from the application or the device or the ambient audio to the machine learning model to determine whether a driving event sound in the audio is artificial.   
     
     
         4 . The method of  claim 3 , further comprising:
 determining, by the one or more processors, that the audio includes the driving event sound.   
     
     
         5 . The method of  claim 4 , wherein determining that the audio includes the driving event sound includes:
 comparing, by the one or more processors, the audio playback data from the application or the device or audio fingerprints included in the ambient audio to one or more audio fingerprints of predetermined driving event sounds.   
     
     
         6 . The method of  claim 4 , wherein the machine learning model is a first machine learning model and further comprising:
 training a second machine learning model using (i) a set of audio streams, and (ii) an indication of the driving event sound corresponding to at least some of the audio streams in the set of audio streams.   
     
     
         7 . The method of  claim 6 , wherein determining that the audio includes the driving event sound includes:
 applying, by the one or more processors, the audio playback data from the application or the device or the ambient audio to the second machine learning model to determine whether the audio includes the driving event sound.   
     
     
         8 . The method of  claim 1 , further comprising:
 providing, by the one or more processors, the trained machine learning model to a client device for the client device to determine whether a driving event sound is artificial.   
     
     
         9 . A server device for training a machine learning model to determine whether a driving event sound is real or artificial, the server device comprising:
 one or more processors; and   a non-transitory computer-readable memory coupled to the one or more processors and storing instructions thereon that, when executed by the one or more processors, cause the server device to:
 obtain a set of driving event sound characteristics for each of a plurality of driving event sounds; 
 for each driving event sound in the plurality of driving event sounds, obtain an indication of whether the driving event sound is from a real or artificial source; and 
 train a machine learning model to determine whether a driving event sound is real or artificial using (i) the set of driving event sound characteristics corresponding to each driving event sound, and (ii) the indication of whether each driving event sound is from the real or artificial source, 
 wherein each set of driving event sound characteristics is classified according to whether the set corresponds to one of the plurality of driving event sounds from the real source or from the artificial source. 
   
     
     
         10 . The server device of  claim 9 , wherein the instructions further cause the server device to:
 obtain audio playback data from an application executing on a client device;   obtain audio playback data from a device communicatively coupled to the client device; or   obtain ambient audio.   
     
     
         11 . The server device of  claim 10 , wherein the instructions further cause the server device to:
 apply the audio playback data from the application or the device or the ambient audio to the machine learning model to determine whether a driving event sound in the audio is artificial.   
     
     
         12 . The server device of  claim 11 , wherein the instructions further cause the server device to:
 determine that the audio includes the driving event sound.   
     
     
         13 . The server device of  claim 12 , wherein to determine that the audio includes the driving event sound, the instructions cause the server device to:
 compare the audio playback data from the application or the device or audio fingerprints included in the ambient audio to one or more audio fingerprints of predetermined driving event sounds.   
     
     
         14 . The server device of  claim 12 , wherein the machine learning model is a first machine learning model and the instructions further cause the server device to:
 train a second machine learning model using (i) a set of audio streams, and (ii) an indication of the driving event sound corresponding to at least some of the audio streams in the set of audio streams.   
     
     
         15 . The server device of  claim 14 , wherein to determine that the audio includes the driving event sound, the instructions cause the server device to:
 apply the audio playback data from the application or the device or the ambient audio to the second machine learning model to determine whether the audio includes the driving event sound.   
     
     
         16 . The server device of  claim 9 , wherein the instructions further cause the server device to:
 provide the trained machine learning model to a client device for the client device to determine whether a driving event sound is artificial.   
     
     
         17 . A non-transitory computer-readable memory coupled to one or more processors and storing instructions thereon that, when executed by the one or more processors, cause the one or more processors to:
 obtain a set of driving event sound characteristics for each of a plurality of driving event sounds;   for each driving event sound in the plurality of driving event sounds, obtain an indication of whether the driving event sound is from a real or artificial source; and   train a machine learning model to determine whether a driving event sound is real or artificial using (i) the set of driving event sound characteristics corresponding to each driving event sound, and (ii) the indication of whether each driving event sound is from the real or artificial source,   wherein each set of driving event sound characteristics is classified according to whether the set corresponds to one of the plurality of driving event sounds from the real source or from the artificial source.   
     
     
         18 . The non-transitory computer-readable memory of  claim 17 , wherein the instructions further cause the one or more processors to:
 obtain audio playback data from an application executing on a client device;   obtain audio playback data from a device communicatively coupled to the client device; or   obtain ambient audio.   
     
     
         19 . The non-transitory computer-readable memory of  claim 18 , wherein the instructions further cause the one or more processors to:
 apply the audio playback data from the application or the device or the ambient audio to the machine learning model to determine whether a driving event sound in the audio is artificial.   
     
     
         20 . The non-transitory computer-readable memory of  claim 19 , wherein the instructions further cause the one or more processors to:
 determine that the audio includes the driving event sound.