Text-to-speech web service Odiogo (Updated)
Update(06/01/2010 10:30 AM): The audio version of this post is now available. You can click the above “listen now” button to listen to this post.
Update (05/31/2010 11:25 AM):Â Apparently, Odiogo hasn’t converted this newest post to audio yet. According to their FAQ, they check the update several times throughout the day. However, it has been almost twelve hours since this post was written, the audio version of this post hasn’t been available yet. I will update this post as soon as the audio version of the post is available. In the meanwhile, you can checkout the TTS version for my other posts on this blog.
While I was reading the news today, I came across an article on freesoftwaremagazin.com. A feature on the site that interests me is the “listen now” button. When I clicked on it, a familiar male TTS (Text-To-Speech) voice began to read the current article aloud for me. This voice is very familiar to me because I always used TextAloud with this Neospeech Paul voice to read the articles online when I was a Windows user. This voice is very human-like compare to Neospeech’s own female voice. After I switched to GNU/Linux for the first time, I tried to install TextAloud and the Paul voice through Wine, but I failed. Because I was too relied on the TTS reading the articles for me, my reading speed have become slow. I then decided to challenge myself to read myself since that day when I gave up installing TTS engine and TextAloud onto GNU/Linux. However, I still think that my reading haven’t improve at all. When I heard this voice on the website, the above memory came up in my mind. I can’t help but to look for where to produce this voice. Fortunately, it contains an introductory voice in the audio that talks about the source of this feature. It’s odiogo.com. At first, I thought that this web site provides web services that can use TTS on other websites to help me read the articles just like TextAloud. However, it seems that it used to convert my own blog to podcasts. It disappoints me to know that it can’t work on websites other than my own site, but I think that it is a good idea to provide more accessibility to my blog through this web service. Therefore, I added a “listen now” button onto this blog to provide text-to-speech feature. Check it out now below the title of this post and tell me what do you think in the comment.
The Format of .desktop files on GNU/Linux
The usage of .desktop files on GNU/Linux is similar to shortcuts on Windows. It places on the desktop as well as in the Application/Start menu for easier access to the installed applications on the hard drive. But unlike shortcuts on Windows are binary files, .desktop files are scripts. They can edited in VIM or gEdit. The format of each line in a .desktop file is key=value. The following elements can be used inside a .desktop file:
- Type (Required)
- Name (Required)
- Exec (Required)
- Version
- Encoding
- GenericName
- NoDisplay
- Comment
- Icon
- Hidden
- OnlyShowIn
- NotShowIn
- Path
- Terminal
- MimeType
- Categories
I also noticed that Comment element has multilingual support. For instance, Comment[ja]=Japanese comment, Comment[pt_BR]=Brazilian Portuguese comment, and Comment[zh_CN]=Simplify Chinese comment. The default language of Comment is English.
To learn more about .desktop files and the description for each element, please read Anatomy of a .desktop File (the above description of Comment element is not in this article).