HELP SAVE STREAMING RADIO - This application seeks to patent a system for discovering new radio stations and podcasts online! Help narrow US patent applications before they become patents HERE

QUESTION - Have you seen anything that was published before Nov, 2008 that discusses:

  1. Creating a list of Internet radio station programs; and
  2. dynamically categorizing this list of programs by mapping a dynamically identified topic to a content classifier;

If so, please submit evidence of prior art as an answer to this question. Only one piece of prior art per answer below. 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 speech-to-text conversion on audio content.

TITLE: System And Method For Internet Radio Station Program Discovery

Summary: [Translated from Legalese into English] A service for discovering Internet radio station programs. A list of Internet radio station programs is obtained. Each program on the list is categorized by mapping a topic to a program using a content classifier. User can discover new programs of interest from the list of programs based on the categories.

  • Publication Number: US 20130110502 A1
  • Assignee: Lemi Technology, LLC
  • Prior Art Date: Seeking prior Art predating November, 2008
  • Open for Challenge at USPTO: Open through Nov 2, 2013
  • Note: This application is a continuation of US Patent 8,359,192. The applicant is attempting to “improve” the claims with this child patent by writing new (evidently broader) claims while retaining the original 2008 priority date. So prior art should predate November, 2008.

Claim 1 requires each and every step below:

A method comprising:

  1. Obtaining a plurality of Internet radio station programs;

  2. For each Internet radio station program of the plurality of Internet radio station programs, dynamically categorizing the Internet radio station program by mapping a dynamically identified topic of the Internet radio station program to a content classifier; and

  3. Enabling a user to discover an Internet radio station program of interest from the plurality of Internet radio station programs based on the dynamic categorizations for the plurality of Internet radio station programs.

In English this means:

  1. Creating a list of internet radio station programs
  2. “Dynamically categorizing” this list of programs by mapping a “dynamically identified topic” to a content classifier; and
  3. Allowing the user to discover programs from the list based on the “dynamic categorization” (presumably created by the classifier)

Good prior art would be evidence of a system that did each and every one of these steps prior to the November, 2008.

You're probably aware of ten pieces of art that meet this criteria already... separately, the applicant is claiming a method using all of the steps above and including speech-to-text conversion to categorize the radio station programs. Extra credit would include systems that do all of the steps above and also involve speech-to-text conversion.

"Internet Radio Station program discovery service" flow chart 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.

6 Answers 6


At audionet.com we categorized radio stations, shows, networks, all content we hosted. We had news, sports, etc. We had source, host, all types if meta data for categories.

In 1998 for broadcast.com we added index.broadcast.com. It indexed all the meta data and where we were able to do speech to text, the actual transcripts of the content.

Any consumer could go to the site and search for any word or term. We used indexing technology from mediasite.com, I believe it was.

  • You didn't mention if it had dynamic classification.
    – George White
    Jun 16, 2013 at 19:01

I did a little searching and found a patent that does not seem to have been cited in this or the parent application. It categorizes text into topics and mentions it being fed by a speech-to-text converter. It mentions radio but not Internet radio.

US618531 - Topic indexing method

Publication date Feb 6, 2001

Filing date Jan 9, 1998

Priority date Jan 9, 1997

This invention generally relates to automatically relating a story to a number of topics. Herein “story” is defined to include any article of information that may be presented to the inventive method, usually from a stored file. More particularly, the present invention relates to automatically indexing and associating stories to a concurrent number of topics ranging from broad to very detailed topics. The present invention is particularly well suited to be combined with voice to text converters, optical readers, and fax encoded data wherein the story (text) is automatically indexed to one or more topics or categories.


Kerbango which launched back in 2000 was the first Internet Radio appliance to my knowledge the systems of discovery and tracking of Internet Radio were launched with the company and most of the application and methods were considered obvious to the product team thus becoming part of the public domain as they didn't have enough capital to file for all of the potential patents in the early space.

The company was purchased by 3Com just prior to the Dot Com bust - 3Com eventually sold the technology to Apple where it currently is still found behind itunes.

The CTO is Carl Hewitt (yes formerly of Apple) - still lives in the San Jose area - Most of what I have read of this patent is a rehash of the same methods used in 2000 for Kerbango - much is documented and found on the Wayback Machine

  • Where do I read about Kerbango classifying channels by dynamic speech-to-text with the text then analyzed to determine a topic classification?
    – George White
    Jun 4, 2013 at 21:10
  • Here is a list of patents in this range from back in 2004 sumobrain.com/… WHen you add prior art from Nuance for the speech component you have a pretty complete set of prior art examples that should help narrow the patent claims.
    – msh317
    Jun 5, 2013 at 14:58
  • Specifically Patent US7469283 sounds very similar sumobrain.com/patents/us/Streaming-media-search-playback-system/…
    – msh317
    Jun 5, 2013 at 15:08
  • The two comments should probably be moved to be answers rather than comments.
    – George White
    Jun 5, 2013 at 16:47

Broadcast.com was started in 1995 by Mark Cuban. Does that fall under this patent?

Also, probably not going far back enough - but I was involved in a radio program in Dallas from 1998 to 2000 that broadcast on the radio, TV and shared audio and video versions of the program online. The servers are no longer there but The Wayback Machine has a cache of the site from 1999 - http://web.archive.org/web/19991014012947/http://www.lighthouse21.com/real.htm


Patent No.: US 7,720,835 B2

Systems and methods for consumer-generated media reputation management

Filed: May 7, 2007

Prior Publication Data

US 2007/0294281 Al Dec. 20, 2007

col 1 ln 59-63

As used herein, the term "Content" refers to media that resides on CGM sites. CGM is often text, but includes audio files and streams (podcasts, mp3, streamcasts, Internet radio, etc.

col 10 ln 44-49

In one embodiment, TruCast is designed to determine, with a high degree of confidence, the conceptual relevance of a given piece of CGM content to a "topic" or concept space. Topics can be of any breadth ("War" is just as sufficient a topic as "2002 Chevy Silverado Extended Cab Door Hinge Bolt Rust").

col 12 ln 56-60

TruCast, in one embodiment can leverage Optical Character Recognition (OCR) image to text conversion, Fast-Fourier Transform (FFT) and Granular Synthesis (GS) analysis based speech-to-text conversion, as well as Hierarchical Temporal Memory (HTM) processing. This comparison, and the resultant threshold filtered probability that a given post is relevant to a given topic allows TruCast, in one embodiment, to assign this meta-information.

col 11 ln 62-67

Topics and keyphrases are created by users (455) in the Topic Manager panel (460) within the VI. Once a new topic and keyphrase is created, this information is transmitted to the indexing system (465) which begins to examine all incoming data for matches against this keyphrase.

It is 54 pages of specification and 40 sheets of drawings. If there is anything else needed it is probably in here somewhere.


Back in 1994-97 almost every college kid at USC had an internet radio station via realNetworks (or a previous incarnation of them).

I personally did 3 shows on my website. However I have no records or files remaining, so this is anecdotal evidence at best.

  • The question is asking for prior art, which must be a document with a verifiable publication date. Anecdotal evidence isn't helpful.
    – stharward
    Jun 8, 2013 at 0:27
  • I remember finding audio streams by music genre on en.wikipedia.org/wiki/SHOUTcast well before 2008 Jul 22, 2013 at 19:32

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .