Automated television program editing
AssigneeKill the Cann, LLC
InventorsKeith Gilbert, Geoffrey F. Gross, Joel MacAuslan
Life of this patent
Estimating the term…
Abstract
The available audio tracks of a multi-track television program are processed to remove or reduce certain components of the program, especially audience sounds, such as laughter. Such processing may be particularly applicable to “canned” laughter. This applies to any program which is supplied with a plurality of audio channels, whether provided by cable, DVD, Internet streaming, or otherwise.
Claims24
A method performed by at least one computer processor, the method comprising:
- (1) receiving a signal representing a plurality of audio channels, wherein the plurality of audio channels includes an audience channel and a performance channel;
- (2) identifying at least one region, in the performance channel, in which sounds of performers and sounds of an audience are present;
- (3) identifying at least one region, in the performance channel, in which sounds of the audience are audible and sounds of the performers are not audible;
- (4) reducing sounds of the audience in the audience channel and the performance channel in the regions identified in (2) and (3), thereby producing a reduced audience channel and a reduced performance channel.
A non-transitory computer-readable medium comprising computer program instructions executable by at least one computer processor to perform a method, the method comprising:
- (1) receiving a signal representing a plurality of audio channels, wherein the plurality of audio channels includes an audience channel and a performance channel;
- (2) identifying at least one region, in the performance channel, in which sounds of performers and sounds of an audience are present;
- (3) identifying at least one region, in the performance channel, in which sounds of the audience are audible and sounds of the performers are not audible;
- (4) reducing sounds of the audience in the audience channel and the performance channel in the regions identified in (2) and (3), thereby producing a reduced audience channel and a reduced performance channel.
A method performed by at least one computer processor, the method comprising:
- (1) receiving a signal representing a plurality of audio channels, wherein the plurality of audio channels includes:
- an original music/sound effect track;
- an audience track;
- and
- a performance track;
- (2) filtering the original music/sound effect track to produce a filtered music/sound effect track;
- (3) subtracting the filtered music/sound effect track from the audience track to produce an audience estimate signal;
- (4) filtering the audience estimate signal to produce a filtered audience estimate signal;
- (5) subtracting the filtered audience estimate signal from the performance track to produce a deflated performance track;
- (6) applying a windowing function to the audience estimate signal to isolate at least one active audience region in the audience estimate signal;
- (7) applying the windowing function to the deflated performance track to isolate at least one active audience region in the deflated performance track;
- and
- (8) applying a time-frequency mask to the isolated active audio regions isolated in (6) and (7) to produce an estimated performance signal.
A non-transitory computer-readable medium comprising computer program instructions executable by at least one computer processor to perform a method, the method comprising:
- (1) receiving a signal representing a plurality of audio channels, wherein the plurality of audio channels includes:
- an original music/sound effect track;
- an audience track;
- and
- a performance track;
- (2) filtering the original music/sound effect track to produce a filtered music/sound effect track;
- (3) subtracting the filtered music/sound effect track from the audience track to produce an audience estimate signal;
- (4) filtering the audience estimate signal to produce a filtered audience estimate signal;
- (5) subtracting the filtered audience estimate signal from the performance track to produce a deflated performance track;
- (6) applying a windowing function to the audience estimate signal to isolate at least one active audience region in the audience estimate signal;
- (7) applying the windowing function to the deflated performance track to isolate at least one active audience region in the deflated performance track;
- and
- (8) applying a time-frequency mask to the isolated active audio regions isolated in (6) and (7) to produce an estimated performance signal.
Dependencies are inferred from the claim text.