Automatic detection, extraction, and memorialization of media content
Communications over a network, such as electronic conferences, often have a small portion of content that is notable within the entirety of the communications. By automatically identifying and preserving such notable portions, the relevant “nugget” from the communications may be maintained without requiring preserving the entirety of the communications. Preservation may comprise converting the notable portion to a non-fungible token and adding the token to a blockchain wherein it may be exchanged by the owner to an interested party.
1 . A system for memorializing a notable portion of a communication, comprising:
a network interface to a network;
a processor; and
a computer-readable medium coupled to the processor, the computer-readable medium comprising one or more computer-readable instructions, the processor executing the one or more computer-readable instructions to:
receive, via the network interface, a communication comprising communication content being exchanged between a plurality of communication devices over the network;
automatically preserve the notable portion in response to identifying the notable portion; and
terminate receiving the communication upon determining the communication is not designated as eligible for identifying notable portions therein.
2 . The system of claim 1 , wherein identifying the notable portion comprises the processor providing the communication content to a neural network trained to determine notable portions therefrom and receiving indicia of the notable portion from the neural network.
3 . The system of claim 2 , wherein the neural network is trained via a computer-implemented method, comprising:
collecting a set of prior communication content from a database;
applying one or more transformations to each set of prior communication content including altering a sentiment, altering a volume of an audio portion of the communication content, altering a level of animation of images of participants in a video portion of the communication content, inserting a change in topic, removing a change in topic, adding a negating word, adding an achievement word, removing the achievement word, adding an achievement image, removing the achievement image, adding an achievement pattern of speech, removing the achievement pattern of speech, removing the negating word, adding a stop word, or removing a stop word to create a modified set of prior communication content;
creating a first training set comprising the set of prior communication content, the modified set of communication content, and a set of non-notable communication content;
training the neural network in a first stage using the first training set;
creating a second training set for a second stage of training comprising the first training set and the set of non-notable communication content that are incorrectly detected as notable portions after the first stage of training; and
training the neural network in the second stage using the second training set.
4 . The system of claim 2 , wherein the notable portion is delineated from an entirety of the communication content as a start of a topic discussed in the communication content that comprises the notable portion.
5 . The system of claim 2 , wherein the notable portion is delineated from an entirety of the communication content as an end of a topic discussed in the communication content that comprises the notable portion.
6 . The system of claim 1 , wherein the processor, upon determining the communication is designated as eligible for identifying the notable portion, enables at least one ancillary communication service to the communication.
7 . The system of claim 1 , wherein the processor preserves the notable portion, further comprising the processor executing the one or more computer-readable instructions to create a non-fungible token of the notable portion.
8 . The system of claim 7 , wherein the processor preserves the notable portion, further comprising the processor executing the one or more computer-readable instructions to add the non-fungible token to a blockchain.
9 . A method memorializing a notable portion of a communication, comprising:
receiving, via a network interface, a communication comprising communication content being exchanged between a plurality of communication devices over a network; and
automatically preserving the notable portion in response to identifying the notable portion; and
wherein upon determining the communication is designated as eligible for identifying the notable portion, enabling at least one ancillary communication service to the communication, wherein the ancillary communication service comprises at least one of recording, whiteboarding, and co-browsing.
10 . The method of claim 9 , wherein identifying the notable portion comprises providing the communication content to a neural network trained to determine notable portions therefrom and receiving indicia of the notable portion from the neural network.
11 . The method of claim 10 , wherein the neural network is trained via a computer-implemented method, comprising:
collecting a set of prior communication content from a database;
applying one or more transformations to each set of prior communication content including altering a sentiment, altering a volume of an audio portion of the communication content, altering a level of animation of images of participants in a video portion of the communication content, inserting a change in topic, removing a change in topic, adding a negating word, adding an achievement word, removing the achievement word, adding an achievement image, removing the achievement image, adding an achievement pattern of speech, removing the achievement pattern of speech, removing the negating word, adding a stop word, or removing a stop word to create a modified set of prior communication content;
creating a first training set comprising the set of prior communication content, the modified set of communication content, and a set of non-notable communication content;
training the neural network in a first stage using the first training set;
creating a second training set for a second stage of training comprising the first training set and the set of non-notable communication content that are incorrectly detected as notable portions after the first stage of training; and
training the neural network in the second stage using the second training set.
12 . The method of claim 10 , wherein the notable portion is delineated from an entirety of the communication content as a start of a topic discussed in the communication content that comprises the notable portion.
13 . The method of claim 10 , wherein the notable portion is delineated from an entirety of the communication content as an end of a topic discussed in the communication content that comprises the notable portion.
14 . The method of claim 9 , wherein receiving the communication content further comprising terminating receiving of the communication upon determining the communication is not designated as eligible for identifying the notable portion therein.
15 . The method of claim 9 , wherein preserving the notable portion further comprises creating a non-fungible token of the notable portion.
16 . The method of claim 15 , wherein preserving the notable portion further comprises adding the non-fungible token to a blockchain.
17 . A system, comprising:
means to receive, via a network interface, a communication comprising communication content being exchanged between a plurality of communication devices over a network;
means to automatically preserve a notable portion in response to identifying the notable portion;
wherein the means to identify the notable portion comprise means to provide the communication content to a neural network trained to determine notable portions therefrom and receive indicia of the notable portion from the neural network; and
means to terminate receiving the communication upon determining the communication is not designated as eligible for identifying notable portions therein.
18 . The system of claim 17 , wherein the neural network is trained via a computer-implemented method, comprising:
collecting a set of prior communication content from a database;
applying one or more transformations to each set of prior communication content including altering a sentiment, altering a volume of an audio portion of the communication content, altering a level of animation of images of participants in a video portion of the communication content, inserting a change in topic, removing a change in topic, adding a negating word, adding an achievement word, removing the achievement word, adding an achievement image, removing the achievement image, adding an achievement pattern of speech, removing the achievement pattern of speech, removing the negating word, adding a stop word, or removing a stop word to create a modified set of prior communication content;
creating a first training set comprising the set of prior communication content, the modified set of communication content, and a set of non-notable communication content;
training the neural network in a first stage using the first training set;
creating a second training set for a second stage of training comprising the first training set and the set of non-notable communication content that are incorrectly detected as notable portions after the first stage of training;
training the neural network in the second stage using the second training set; and
preserving the notable portion, further comprising creating a non-fungible token of the notable portion.
19 . The system of claim 18 , further comprising means to add the non-fungible token to a blockchain.
20 . The system of claim 17 , further comprising means to, upon determining the communication is designated as eligible for identifying the notable portion, enable at least one ancillary communication service to the communication.