Back to results
US 2026/0301731 A1Published application
Systems and methods for ai-driven contextualized multi-cast narration
OwnerPhilip Dana Marshallinventor
Life of this patent
Estimating the term…
Abstract
Systems and methods are herein provided for an audio narration system. In one example, an audio narration system comprises a processor communicably coupled to non-transitory memory storing one or more AI models, the non-transitory memory including instructions that when executed cause the processor to: receive text data from a user input device; process the text data with the one or more AI models; generate an audio narration with the one or more AI models; and output the audio narration to the user input device.
Claims20
An audio narration system, comprising:
- a processor communicably coupled to non-transitory memory storing one or more AI models, the non-transitory memory including instructions that when executed cause the processor to:
- receive text data comprising a plurality of passages from a user input device;
- process the text data with the one or more AI models to determine text classifications of each passage of the plurality of passages of the text data and one or more context parameters;
- generate an audio narration with the one or more AI models based on the text classifications of each passage of the plurality of passages and the one or more context parameters, wherein a delivery of a given passage in the audio narration reflects the text classification thereof and the one or more context parameters of the given passage;
- and
- output the audio narration to the user input device.
3 more elements
A method for an audio narration system, comprising:
- receiving text data, wherein the text data comprises a plurality of passages;
- determining text classifications for the plurality of passages, wherein each passage of the plurality of passages is classified as one of dialogue, narration, and a meta-narration element;
- determining one or more context parameters of the text data, wherein the one or more context parameters are determined for each passage of the plurality of passages and one or more of the context parameters are determined for sequences of passages of the plurality of passages;
- generating an audio narration based on the text classifications and the one or more context parameters. wherein the audio narration retains narrative context, character identity, and delivery attributes across passages based on the text classifications and the one or more context parameters;
- and
- outputting the audio narration to a user device.
3 more elements
An audio narration system, comprising:
- a local processing component disposed at an edge device, the edge device comprising non-transitory memory storing a text analysis model and instructions that when executed cause a first processor of the edge device to:
- receive text data from a user input device communicatively coupled to the edge device;
- and
- process the text data with the text analysis model determine, locally at the edge device, text classifications for each passage of the text data and one or more context parameters for one or more of each passage of the text data and sequences of passages of the text data;
- and
- a text narration model stored remotely in a server, wherein the text narration model is configured to ingest the text data, the text classifications for each passage of the text data, and the one or more context parameters;
- and
- generate an audio narration of the text data based on the text classifications for each passage of the text data and the one or more context parameters;
- and
- transmit the audio narration to the edge device for output via the user input device.
7 more elements
Dependencies are inferred from the claim text.