Fubby-Chia Connections
I’ve been thinking of doing something that might add to the value of this page for a long time. One of my first thoughts was to do a name index. Then, I remembered an old suggestion given by our friend K9 to podcast. That way, people could listen to them anywhere, anytime. Perhaps they could use them to pass the time on long car trips, or play them late at night if they really needed to get to sleep fast.
Well, I wouldn’t wanna read all that. And heaven knows I can’t pay anyone to read it for me. So I thought about some kind of text-to-speech technology. Last year, I signed up with odiogo.com, because they had a program that automatically translated blogs to TTS podcasts.
Supposedly.
I had a number of problems with odiogo.com. First of all, because of the design and programming of this page, it won’t work. So, I created another blog where it could. I then found out that the podcasts only work temporarily. And downloading them (the whole point of the exercise) proved cumbersome.
Then, there’s something else that’s common to all TTS software. They’re difficult to edit. And lord knows they need lots of editing. They doth murder proper nouns. Poor innocent proper nouns. Especially those that are not English in origin. They also have a tendency to screw up acronyms. And this page is full of them. Oftentimes, they would pronounce a phrase such as “Dr. Timothy Leary” as “Drive Timothy Leary.” Just as often, the phrase “FBI-CIA connections” came out as “Fubby-Chia connections.”
Odiogo doesn’t allow for editing, unfortunately. So I began to consider purchasing a TTS software that does. I’m currently giving Ivona [http://www.ivona.com/en/] a try, courtesy of its thirty-day free trial period. The posts still take a lot of editing to correct pronunciation, and make them sound a little less as though they’re read by robots.
So, before I actually go through the time and expense of purchasing software and editing complete series, I thought I’d ask for your honest opinion. If you can listen to one, or two, or all of them, I’d appreciate it if you could tell me (a) if they’re understandable; (b) if they’re something you’d want to listen to, or if you know someone who’d want to listen to them; and (c) if they’re worth the effort.
Click on the titles to download, or just play them here.
1. The Lurking Porno Boogeyman
2. Turn Me On, Dead Man
3. ...And the Superstar, Pt. I & II
4. The Devil’s in the Slide: Intelligence Design, Pt. I
5. Self-Funded Commentary on McMartin
6. Death by JFK Assassination
7. The Hoax that Launched a Thousand Ships, Pt. I & II
Well, I wouldn’t wanna read all that. And heaven knows I can’t pay anyone to read it for me. So I thought about some kind of text-to-speech technology. Last year, I signed up with odiogo.com, because they had a program that automatically translated blogs to TTS podcasts.
Supposedly.
I had a number of problems with odiogo.com. First of all, because of the design and programming of this page, it won’t work. So, I created another blog where it could. I then found out that the podcasts only work temporarily. And downloading them (the whole point of the exercise) proved cumbersome.
Then, there’s something else that’s common to all TTS software. They’re difficult to edit. And lord knows they need lots of editing. They doth murder proper nouns. Poor innocent proper nouns. Especially those that are not English in origin. They also have a tendency to screw up acronyms. And this page is full of them. Oftentimes, they would pronounce a phrase such as “Dr. Timothy Leary” as “Drive Timothy Leary.” Just as often, the phrase “FBI-CIA connections” came out as “Fubby-Chia connections.”
Odiogo doesn’t allow for editing, unfortunately. So I began to consider purchasing a TTS software that does. I’m currently giving Ivona [http://www.ivona.com/en/] a try, courtesy of its thirty-day free trial period. The posts still take a lot of editing to correct pronunciation, and make them sound a little less as though they’re read by robots.
So, before I actually go through the time and expense of purchasing software and editing complete series, I thought I’d ask for your honest opinion. If you can listen to one, or two, or all of them, I’d appreciate it if you could tell me (a) if they’re understandable; (b) if they’re something you’d want to listen to, or if you know someone who’d want to listen to them; and (c) if they’re worth the effort.
Click on the titles to download, or just play them here.
1. The Lurking Porno Boogeyman
2. Turn Me On, Dead Man
3. ...And the Superstar, Pt. I & II
4. The Devil’s in the Slide: Intelligence Design, Pt. I
5. Self-Funded Commentary on McMartin
6. Death by JFK Assassination
7. The Hoax that Launched a Thousand Ships, Pt. I & II
Labels: cyberculture, personal stuff



10 Comments:
At 8:33 PM,
foam said…
Umm, I don't know, xdell.I might listen to them while driving in the car. I would rather read your articles though. The voices do sound a bit odd. Your voice would sound wonderful though.
And this is what my almost 15 year old had to say .... What's that, he asked while I was listening. Something like an audiobook, he continued? Yeah, something like that, I answered. Whaddaya think.? Would you listen to it? No, he said. It sounds awful.
At 10:20 PM,
X. Dell said…
Foam, I won't say anything right now, except thanks for listening.
At 6:53 AM,
Shrinky said…
I like the idea of listening, as opposed to reading, as your subjects do require a longer period of focus and time, to grant them full justice (as opposed to the usual attention span required with most blog-posts). A pod-cast can be heard and digested without necessitating the listener to "down tools" and dedicate themselves solely to that one purpose. I would happily engage in listenning to this as I pottered around, say cooking, or driving.
That said, the voices are awkward, some more stilted than others, and lack the true intonation I assume should be present. I think Foam has a valid point in that it feels clear something sounds "wrong" there, it distracts from the message, and I found myself straining at times to interperet certain points.
Wouldn't the ideal here, be for you to record your own voice? There is no passion nuance, or conviction in these tapes - you, as the author, can inject this, making it far more easier for us to engage.
At 7:42 AM,
J Cosmo Newbery said…
Watch this space - down loaded one, will transfer to my phone and play while going to work on Thursday. (It's my listening time.)
PS: Word Verification tests my patience.
At 10:53 AM,
Charles Gramlich said…
My kindle does the same thing when I let it read to me. I had the word "Pa" in a book once and the Kindle kept translating it as Pennsylvania.
At 11:06 AM,
X. Dell said…
Shrinky, thanks for the input. I have reasons for not reading it myself--although that would be quicker and easier, and probably sound more human. Perhaps at one point I'll consider that more strongly. I have to weigh a lot of things in my head before I do that.
Cosmo, thanks. That will be a test. Also, I'm with you about word verification. I never liked it in the first place, and for years I managed to get by without it. I guess the experience of erasing spam after spam (one at a time, no less) changes one's outlook a bit. But I noticed that the latest version of Blogger has a spam guard. Maybe, I'll give THAT a whirl.
Lol, Charles. Good thing you weren't listening to a history of Massachusetts and Pennsylvania Kettle.
At 6:20 PM,
Ray Palm (Ray X) said…
(A) Understandable but really annoying to me. Details below.
(B) No, unless I was blind and there was no other way to get the info.
(C) No. Can’t you just use your god-given voice? Even if you make mistakes, I can let that pass more than trying to strain my ears listening to a robo-voice.
In case you didn’t know the government runs regional radio stations that provide local weather information – the “weather band” on some radios or “WX” on a police scanner. The station near here changed to the TTS format and it was annoying as hell. The voice has improved somewhagt since then but there are other problems as you mentioned.
For example, there was a message between forecasts about a meeting to train weather watchers to be held on “December 2 North Dakota.” Whoever typed in “December 2nd” didn’t bother to fix the mistake. GIGO: Garbage In, Garbage Out.
Besides automated errors like that, what is the value of if mistakes like this one aren’t reviewed and corrected:
NOAA Weather Station, Vermont: “3 AM regional conditions… Plattsburgh, NY – sunny.”
With the change over to broadcast digital TV I now can access a dedicated weather channel, a subchannel of a local station. I don’t bother with hitting the WX button on my scanner. Yes, the forecasters might flub their lines but it’s easier than straining my ears.
With all the effort you put into playing around with TTS software, you could have spent the time writing. Also your posts are “dense” in the sense of having a lot of information, the way you reason out certain points to draw conclusions. Reading is a better way to absorb the info.
Ma and Pa Kettle say hello.
At 6:22 PM,
Ray Palm (Ray X) said…
That should be "The voice improved somewhat..."
At 11:00 AM,
Dr.Alistair said…
good god man.....those voices are vile!
a natural voice, mistakes and all, would be much preferred to that.
it's like listening to singers who have been autotuned...
if you have a moment, take a look at some of my new artwork at hypgnosys...and tell me what you think.
cheers, A.
At 12:25 PM,
X. Dell said…
Okay, Ray. Thanks for the imput.
You too, Dr. A. Will get to your site anon.
Post a Comment
Linkbacks:
Create a Link
<< Home