AU Interactive

What’s your opinion of podcasting?

I was thinking about my blogging pattern and I realized that I’m probably better at speaking than I am at writing. I prefer the stream-of-consciousness approach. I’ve never been a big fan of podcasting on the receiving end because I don’t have to commute to work and I hate iTunes. But I realize there is a market for it and some people swear by podcasts. My biggest issue with podcasting vs. blogging is that it’s (possibly) good content that cannot be indexed by the search engines so you have to rely on spreading the message virally and really miss out on the search engine optimization angle.

So I thought about podcasting transcription and ran a few searches to see if such a service exists. It looks like CastingWords offers such a service (at 42 cents/minute). There are some others but they charge upwards of $3-4/minute which is way too expensive. It seems that to do it efficiently you could run the audio through something like Naturally Speaking (voice recognition software), then have a human edit the raw output. The services I’ve seen all look like they’re completely human powered (which doesn’t seem too efficient). If I was to do podcasting, I’d like it to be transcribed as well - so I can double dip on the content AND allow people to have the option to read it.

Any feedback on whether I should experiment with podcasting? Would anybody care to hear my thoughts about online trends, what’s working and what’s new in the web 2.0 world, development, new services, etc - maybe once a week?

Gareth said,

November 3, 2006 @ 6:25 pm

I’d rather read the content than listen to it. That way I can go at my own pace, and absorb the information.

Rewinding and relistening just isn’t the same as reading the same sentence over a few times until it sinks in.

Richard said,

November 4, 2006 @ 11:27 am

This seems to be exactly the kind of service that Amazon’s Mechanical Turk would be great at. I’ve seen reports that people end up working for ~ $0.10/minute doing these kind of services, simply because they like “sticking it to the man” by earning extra money in a weird way.

Yeah. Weird.

I wonder how well this could really be outsourced, considering that the average podcast probably contains a large number of very current topical references that might be hard to transcribe “blind,” if you know what I mean.

Devin Voorsanger said,

November 7, 2006 @ 2:29 am

If you are actually able to get Naturally Speaking trained to the point where it’s not more of a pain in the “tukus” to recite than it is to just type it, I would want both. That way I could choose how I wanted to consume it(and isn’t that the whole point anyway)

The one thing I would be worried about is that with the structure of the written word (the beat poets aside) is very different than the structure of the spoken word. The written word by it’s very design is for communicating with strangers. It’s designed to have a formal structure that multiple people who don’t have your cultural references or accent or slang will still be able to follow. Video would be interesting if you make it interesting for that medium. This means not just another talking head but a little bit of your personality like Bill Pogue does with his NYT digital podcast.

On search within the video: Klipmart out of Doublclick and Atlas both have interaction and tracking of interaction within the video. With Google’s acquisition of Youtube you should see contextual and semantic ad tracking within the video in 1 year and they will then index all your old podcasts and you will be richly rewarded as long as their still relevant.

Markus said,

November 7, 2006 @ 9:48 am

I agree that the structure is quite different - that’s a very good point - but I think you might be able to get away with it if you disclaim it as a transcript up front.

Leave a Comment