Automated Audio Mapping Using an Artificial Neural Network

Number: 20210099760
Published: 2021-04-01
Filed: 2019-09-27
Assignee: Disney Enterprises, Inc.
Inventors: Farre Guiu; Miquel Angel, Martin; Marc Junyent, Aparicio; Albert, Swerdlow; Avner, Accardo; Anthony M., Anderson; Bradley Drew
CPC: G10L15/005; G10L25/81; H04N21/4666; G10L25/54; H04N21/4394; H04N21/8106; G06N3/08; G10L25/30; G10L25/18; G10L25/51; G06N3/09; G10L15/16
Verdict: Set aside audio-content mapping ML tool, business/generic
Source: Google Patents · FreePatentsOnline

Abstract

According to one implementation, an automated audio mapping system includes a computing platform having a hardware processor and a system memory storing an audio mapping software code including an artificial neural network (ANN) trained to identify multiple different audio content types. The hardware processor is configured to execute the audio mapping software code to receive content including multiple audio tracks, and to identify, without using the ANN, a first music track and a second music track of the multiple audio tracks. The hardware processor is further configured to execute the audio mapping software code to identify, using the ANN, the audio content type of each of the multiple audio tracks except the first music track and the second music track, and to output a mapped content file including the multiple audio tracks each assigned to a respective one predetermined audio channel based on its identified audio content type.

Background

BACKGROUND

Audio-visual content, such as movies, television programming, and broadcasts of sporting events, for example, is widely used to distribute entertainment to consumers. Due to its popularity with those consumers, ever more audio-visual content is being produced and made available for distribution via traditional broadcast models, as well as streaming to services. Consequently, the accuracy and efficiency with which audio-visual content can be reviewed, classified, archived, and managed has become increasingly important to producers, owners, and distributors of such content.

Effective management of audio-visual content includes the classification of audio tracks accompanying the video assets contained in an audio-visual file. However, conventional approaches to performing audio classification tend to be intensively manual, making those conventional processes relatively slow and costly. As a result, there is a need in the art for solutions enabling automation of the classification of audio tracks accompanying the video components of audio-visual content, in order to reduce the time and human involvement required for content classification and management. SUMMARY

There are provided systems and methods for performing automated audio mapping using an artificial neural network, substantially as shown in and/or described in connection with at least one of the figures, and as set forth more completely in the claims.

Claims

1. An automated audio mapping system comprising: a computing platform including a hardware processor and a system memory; an audio mapping software code stored in the system memory, the audio mapping software code including an artificial neural network (ANN) trained to identify a plurality of different audio content types; the hardware processor configured to execute the audio mapping software code to: receive a content including a plurality of audio tracks; identify, without using the ANN, a first music track and a second music track of the plurality of audio tracks; identify, using the ANN, an audio content type of each of the plurality of audio tracks except the first music track and the second music track; and output a mapped content file including the plurality of audio tracks each assigned to a respective one of a plurality of predetermined audio channels based on its identified audio content type. || 11. A method for use by an automated audio mapping system including a computing platform having a hardware processor and a system memory storing an audio mapping software code including an artificial neural network (ANN) trained to identify a plurality of different audio content types, the method comprising: receiving, by the audio mapping software code executed by the hardware processor, a content including a plurality of audio tracks; identifying, by the audio mapping software code executed by the hardware processor and without using the ANN, a first music track and a second music track of the plurality of audio tracks; identifying, by the audio mapping software code executed by the hardware processor and using ANN, an audio content type of each of the plurality of audio tracks except the first music track and the second music track; and outputting, by the audio mapping software code executed by the hardware processor, a mapped content file including the plurality of audio tracks each assigned to a respective one of a plurality of predetermined audio channels based on its identified audio content type.