Data Mining: Text Mining, Visualization and Social Media

Feed2Podcast Rocks

After writing about, but failing to implement, a podcast based on my blog using Feed2Podcast, the Feed2Podcast folks quickly identified my error and provided me with a fix. Way to go!

New Podcasting Transcripting Service: CastingWords

Business2 posts about a podcasting transcription service: CastingWords:

There are hardly any podcast transcripts in its search engine so far, and its own site isn't even operational yet. But the idea is so interesting (Michael Arrington of TechCrunch would agree) that I contacted the company to find out more. It turns out to be even more whacky than I originally thought because the company wants to use Amazon's Mechanical Turk service to get people to do the transcribing. One of the founders, Nathan McFarland, explains what the company is hoping to do:

We're a super early stage startup -  the idea is to sell search ads against the keywords in the transcriptions, which we generate using Amazon's  Mechanical Turk ( http://mturk.com) and some whizzy software.   But we need capital to pay all those MTurk workers, so in the next few weeks we'll be opening a store.  It will allow Podcasters to purchase transcriptions of their shows. We'll  do the transcriptions and give them a full transcription - not just chunks pulled back by the search engine if it happens to index their show. Of course they automatically  get listed in the engine, so this arrangement should drive traffic to them, get them transcripts, and get us the cash needed to keep on transcribing...

The post is worth a read.

Radio with Pictures

Moments after I post about why the new iPod is not the new television, but could well be the next radio, Rubel posts this:

Billboard reports that iPod-friendly Rush Limbaugh video clips are on the way. Meanwhile, Broadcasting and Cable says Disney will roll out a massive effort to use iPods to promote the upcoming film The Chronicles of Narnia: The Lion, The Witch and The Wardrobe.

Auto-podcasting, Part 2

I posted recently about using text to speech synthesis to automate podcasting. Rebotcast does exactly that for backstage  (aka BBC) feeds. The site was pointed out on the backstage mailing list by Ted Gilchrist (Botcast Network). It seems that he has been doing this since at least October 6th 2004.

Auto-podcasting

One of the potential values of podcasting is drive-/walk-/run-time access to blog posts. However, a good blogger does not necessarily a good radio voice make. I've been playing with Cepstral's text to speech (tts) synthesis demo and found it to be remarkably good at speaking short blog posts. Here is an mp3 (wav) of a recent post.

This demo is of the Cepstral system with no blog specific mods of any type. Certain structures (in this case parenthetical text) could be dealt with in a slightly more listener friendly manner, but as an out of the box demo, this is really great.

Now imagine being able to listen to posts from your morning read away from the computer. Systems like Cepstral's offer multiple voices, and many configurations providing lots of potential to vary the voice per post, have quoted material spoken in a distinct voice and so on.

Note that speech is another example of an annotation stream!