Browse Prior Art Database

Adaptive generation of narrative music from NLP and sentiment/tone analysis

IP.com Disclosure Number: IPCOM000249549D
Publication Date: 2017-Mar-03
Document File: 3 page(s) / 41K

Publishing Venue

The IP.com Prior Art Database

Abstract

This disclosure proposes a system wherein readers of books or web sites could be provided with a dynamic soundtrack that responds to the mood of the text, changing as the reader passes through different emotional passages.This invention is designed to dynamically generate appropriate music to a person reading a text. It comprises of a component that detects where in a given text the reader is at any given moment, a component that reads the text and processes it using NLP and a cognitive engine, to produce mood or tone classification and a component that adaptively generates music, responding to tone/mood modifiers from the NLP/Cognitive component.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 52% of the total text.

1

Adaptive generation of narrative music from NLP and sentiment/tone analysis

As technology advances, the entertainment industry continually looks for ways in which user experience during the consumption of media can be enhanced. The provision of music and sound tracks to all manner of media is well established and valued – e .g. games, movies, tv shows, some web sites all feature music at some point.

Our disclosure proposes a system wherein readers of books or web sites could be provided with a dynamic soundtrack that responds to the mood of the text, changing as the reader passes through different emotional passages. This invention is designed to dynamically generate appropriate music to a person reading a text . It comprises of;

1) A component that detects where in a given text the reader is at any given moment.

2) A component that reads the text and processes it using NLP and a cognitive engine , to produce mood or tone classification and:

3) A component that adaptively generates music, responding to tone/mood modifiers from the NLP/Cognitive component

The invention comprises 3 distinct components, and the flow of data through them.

1) Reading progress detection component – The purpose of this component is to detect the position in a given text that a reader has reached at any given moment. This component could be made using an eye tracking system ( https://en.wikipedia.org/wiki/Eye_tracking ), or by measuring a readers average reading rate, based on page turns and timing, and getting an average expected position. This component tracks where the reader is and continually feeds this cursor position into the NLP/Cognitive component.

2) The NLP/Cognitive component – Reads the “current” portion of text, (i.e. paragraph) up until the current point of reader progress. The granularity of this portion could be changed from small units to big units . Feeds this portion of text into a mood/tone analysis engine, such as can be built using a cognitive engine. Extracts instantaneous mood/tone data. This data (could be key words or a numerical representation) i...