Patent №
US 11,736,775
Granted
2023-08-22
Filed 2021
Owner
AMAZON TECHNOLOGIES, INC.
Lab
AI components
2
nlp · speech
Assignment
Recorded
Dataset
AIPD
2023_r1 edition
Application
17643569
Methods and apparatus are described for generating audio descriptions of live events in near real time. Visual elements are identified in video frames. Semantic representations of the visual elements are determined and used to generate an audio description. The audio description is provided to a client device for playback during the live event as an alternative to the original audio content.
Natural languageSpeechH04N 21/8106G06V 20/42G06V 20/46H04N 21/2187H04N 21/23418H04N 21/234336H04N 21/251H04N 21/26258+1 more
AI classification
Speech1.00
Natural language0.99
Knowledge representation0.24
Evolutionary computation0.13
Machine learning0.05
AI hardware0.03
Vision0.01
Planning0.00
Ownership
AMAZON TECHNOLOGIES, INC.
assignment · 597710009
Assignors
BALEST, ADAM M
On an employer assignment, the assignors are typically the inventors.
From the same owner
CONTEXT-AWARE BIFURCATED ATTENTION FOR EFFICIENT SAMPLINGUS 12,693,835 · 2026USER-CONFIGURABLE OBJECT GENERATIONUS 12,694,012 · 2026SYSTEMS FOR CONTEXT-AWARE NAMED ENTITY RECOGNITIONUS 12,694,216 · 2026IMAGE-BASED TEXT TRANSLATION AND PRESENTATIONUS 12,694,234 · 2026SCRATCHPAD TENSOR ALLOCATIONUS 12,694,296 · 2026Verification of Neural Network Model EvolutionUS 12,694,298 · 2026RECOMMENDATION TECHNIQUES FOR TARGET EXECUTIONUS 12,694,426 · 2026DEEP NEURAL NETWORK TRAINING DATA GENERATIONUS 12,688,414 · 2026