Surety is performing system maintenance this weekend. Electronic date stamps on new Prior Art Database disclosures may be delayed.
Browse Prior Art Database

Voice Data Parsing

IP.com Disclosure Number: IPCOM000234023D
Publication Date: 2014-Jan-07
Document File: 2 page(s) / 26K

Publishing Venue

The IP.com Prior Art Database


The invention describes a method and system for users to quickly sort through a recorded audio file by various sort criteria, including: speaker; key words; time zones; and other criteria. The system analyzes the audio file in its entirety and extracts text or voice pieces to create a summary of the audio file that the user defined. The user can then select which portions of the audio file they want to review.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 64% of the total text.

Page 01 of 2

Voice Data Parsing

The core idea of this invention is to using existing voice and conferencing technology to incorporate a user interface that will allow users to define specific sort criteria for pre-recorded meetings, conferences, webcast or other types of audio files..

One illustration of this invention is:

A person on the audio file is assigned voice text using existing voice


recognition technology

The audio of the person is converted to text


Each text line (defined as a sentence) is tagged with a unique identifier


and speaker

The user determines the search criteria


Metadata (meetings, email, calendar, location, topic, jukebox, speaker, etc)

Speaker from audio file (direct reports, people you follow, org etc)

The system generates and displays a summary in text of the search


criteria results

The user refines and picks lines of interest


The system creates a text and/or audio file of the lines of interest


The example below shows how an audio file could be converted in the background:

[00001] [mmaresh] [151] this is a test of the voice parsing

[00002] [mmaresh] [202] the first bracket is the identifier

[00003] [mmaresh] [183] the second bracket identifies the speaker

[00004] [mmaresh] [323] the third bracket is the length of the pause in centiseconds (cs) after this line is spoken

[00005] [estegner] [210] i understand

[00006] [Jkogut] [437] so do i

The example below shows a search criteria from the converted file above:


Page 02 of...