US 12,070,688 B2
Apparatus and method for audio data analysis
Marina Villanueva Barreiro, Acoruna (ES); Michael Lee Jones, London (GB); Oliver Hume, London (GB); Fabio Cappello, London (GB); and Danjeli Schembri, London (GB)
Assigned to Sony Interactive Entertainment Inc., Tokyo (JP)
Filed by Sony Interactive Entertainment Inc., Tokyo (JP)
Filed on Oct. 12, 2021, as Appl. No. 17/498,947.
Claims priority of application No. 2016305 (GB), filed on Oct. 14, 2020.
Prior Publication US 2022/0111294 A1, Apr. 14, 2022
Int. Cl. A63F 13/54 (2014.01); G06F 40/20 (2020.01); G10L 15/16 (2006.01); G10L 15/22 (2006.01); G10L 25/30 (2013.01); G10L 25/60 (2013.01)
CPC A63F 13/54 (2014.09) [G06F 40/20 (2020.01); G10L 15/16 (2013.01); G10L 15/22 (2013.01); G10L 25/30 (2013.01); G10L 25/60 (2013.01)] 13 Claims
OG exemplary drawing
 
1. A data processing apparatus, comprising:
input circuitry to receive audio data for a plurality of respective dialogue recordings for a video game;
classification circuitry comprising one or more machine learning models to receive at least a portion of the audio data for each dialogue recording and trained to output classification data indicative of a quality classification of a dialogue recording in dependence upon one or more properties of the audio data for the dialogue recording; and
storage circuitry to store identification data for one or more of the plurality of dialogue recordings in dependence upon the classification data, wherein:
the one or more machine learning models comprise a first machine learning model trained to output first classification data indicative of an audio noise quality classification of the dialogue recording in dependence upon one or more first properties of the audio data for the dialogue recording; and
the storage circuitry is configured to store the identification data for one or more of the plurality of dialogue recordings for which the first classification data is indicative of an audio noise quality classification having an audio noise quality level that is less than a threshold audio noise quality level for the first classification data.