Speech Recognition - Searching Conversations by Keyword - Patent Application - PRIOR ART REQUEST

Question

AN OVERBROAD PATENT ON AUDIO SEARCHING BY KEYWORD - This application from Electronics And Telecommunications Research Institute (Korea) seeks to patent the idea of...indexing recording conversation with keywords for search! 10 minutes of your time can help narrow US patent applications before they become patents. Follow @askpatents on twitter to help.

QUESTION - Have you seen anything that was published before Dec 20, 2011 that discusses:

• Recording conversations and indexing them with keywords which can be searched;

If so, please submit evidence of prior art as an answer to this question.. We welcome multiple answers from the same individual.

EXTRA CREDIT - A reference to anything that meets all of the criteria to the question above AND ALSO uses MORPHEMES, ONTOLOGIES or END-POINT DETECTION (EDP).

TITLE: METHOD FOR PROVIDING CONTENTS ABOUT CONVERSATION

Summary: [Translated from Legalese into English] Recording an audio conversation, indexing it with search keywords, searching the conversation by the search keywords.

• Publication Number: US 20130159003 A1
• Application Number: 13/592,615
• Assignee: Electronics And Telecommunications Research Institute (Korea)
• Prior Art Date: Seeking prior Art predating Dec 20, 2011
• Open for Challenge at USPTO: Open through Dec 20, 2013

Claim 1 requires each and every step below:

A method of providing contents about conversation, comprising:

1. a voice information collecting step of collecting voice information from conversation between a user and another person;

2. a keyword creating control step of creating search keywords by using the collected voice information; and

3. a contents providing control step of searching contents by using the created search keywords, and providing the searched contents.

In English this means:

A method for indexing audio conversations:

1. Recording a conversation between a user and another person

2. Creating search keywords from the recorded conversation

3. Allowing user to search the audio conversation based on the keyword and returning search results

Good prior art would be evidence of a system that did each and every one of these steps prior to the Dec 20, 2011.

You're probably aware of ten pieces of art that meet this criteria already... separately, the applicant is claiming using MORPHEMES, ONTOLOGIES or END-POINT DETECTION (EDP)

---

"Providing contents about conversation” from the Applicant

---

What is good prior art? Please see our FAQ.

Want to help? Please vote or comment on submissions below. We welcome you to post your own request for prior art on other questionable US Patent Applications.

---

Answer 1

Paper on keyword search of spoken conversations from 2005. http://acl.ldc.upenn.edu/H/H05/H05-1119.pdf

SEARCHING THE AUDIO NOTEBOOK: KEYWORD SEARCH IN RECORDED CONVERSATIONS Peng Yu, Kaijiang Chen, Lie Lu, and Frank Seide Microsoft Research Asia, 5F Beijing Sigma Center, 49 Zhichun Rd., 100080 Beijing, P.R.C. {rogeryu,kaijchen,llu,fseide}@microsoft.com

from Proceedings of Human Language Technology Conference and Conference on Empirical Methods in Natural Language Processing (HLT/EMNLP), pages 947–954, Vancouver, October 2005. c 2005 Association for Computational Linguistics

Answer 2

Back in 2005, AT&T was already publishing paper about text mining the speech-to-text transcripts of call centers. The process goes actually much further than just plain indexing for search, because it attempts later stage, more complicated processing, such as sort of de-duplication of the redundant content.

Answer 3

There was a company called Pluggd that did something similar for podcasts back in 2006. Here's an article from Wired that describes in layman's terms what they did:

http://www.wired.com/techbiz/it/news/2006/12/72326

The article mentions that the technology was "patent pending" at that time, so it may be worth digging through the Pluggd patents to understand their technology better.