US 12,008,038 B2
Summarization of video artificial intelligence method, system, and apparatus
Vikram Chalana, Bothell, WA (US); Vishal Chalana, Bothell, WA (US); Shailendra Singh, Hyderabad (IN); and Abid Ali Mohammed, Bothell, WA (US)
Assigned to Pictory, Corp.
Filed by Pictory, Corp., Bothell, WA (US)
Filed on Jan. 5, 2022, as Appl. No. 17/569,386.
Claims priority of provisional application 63/133,973, filed on Jan. 5, 2021.
Prior Publication US 2022/0215052 A1, Jul. 7, 2022
Int. Cl. G06F 16/738 (2019.01); G06F 16/75 (2019.01); G06F 16/783 (2019.01); G06N 3/045 (2023.01); G06N 3/08 (2023.01)
CPC G06F 16/739 (2019.01) [G06F 16/75 (2019.01); G06F 16/7844 (2019.01); G06N 3/045 (2023.01); G06N 3/08 (2013.01)] 18 Claims
OG exemplary drawing
 
1. An apparatus for summarizing an audio-visual media, comprising:
a computer processor and a memory; and
an audio-visual media synopsis module in the memory to summarize the audio-visual media, wherein to summarize the audio-visual media, the computer processor is to perform the audio-visual media synopsis module and is to thereby obtain the audio-visual media, perform a transcription neural network to prepare a transcript of the audio-visual media, perform a summarization neural network to prepare a transcript summary of the transcript of the audio-visual media, and wherein the audio-visual media synopsis module is to output the transcript summary, wherein the audio-visual media synopsis module is further to prepare and output a video summary of the audio-visual media, wherein the video summary of the audio-visual media comprises portions of the audio-visual media corresponding to the transcript summary;
wherein the audio-visual media synopsis module is further to provide the summarization neural network with the transcript and the audio-visual media and wherein to prepare the transcript summary, the summarization neural network is to identify a plurality of sentence meaning clusters in the transcript based on the transcript and the audio-visual media and wherein to prepare the transcript summary further comprises to select a number of sentences from the plurality of sentence meaning clusters.