Wikipedia talk:WikiProject Spoken Wikipedia/Archive 3

From Wikipedia, the free encyclopedia

 This is an archive of past discussions. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.
← Archive 2 Archive 3 Archive 4 →

Contents

Lead paragraph only

While looking for a really short article to try as my first spoken wikipedia attempt, I clicked on the calendar of the "tomarrow's featured article" page, which gave me a page containing only the first paragraph of the article. Thinking that it was a very short, article, I read it, edited it, processed it and all. Now I realize my silly mistake. Should I post the file anyway? It is too late for me to finish it in time for its frot page debut.--William Morgan Jan 8, 2006

Sound icon

The speaker icon that appears on pages containing spoken audio links to the entry for the image itself. It would be more user friendly if the image instead linked to the audio file. That would probably be an exception to the normal Wikipedia method of handling images, however the amount of people that would be interested in additional information on the image would be very, very small. --Dan East 16:54, September 8, 2005 (UTC)

The reason you haven't seen many images that link anywhere else (including the icons on the Main Page for Wiktionary etc.) is because it's not possible. It would make it too easy to hide away the image description page; or at least that's the "official" reason for not implementing this feature. — Timwi 21:39, 10 September 2005 (UTC)
na:Template:Click lets you do it. It's not in use on en: though. Jon Harald Søby \ no na 10:21, 4 December 2005 (UTC)

Spoken articles by date

Is there a way to see a list of the spoken articles by the date they were added? If not, I think there should be one because:

  • I (and probably others) would like to have a way to see each new spoken article as it is added
  • Seconded. It would be a sort of podcast thing run off of Wikipedia articles. Each morning you could see which articles have been newly recorded and download the interesting ones to your portable music player and listen on the commute to work. I say "portable music player" because the vast majority of them (such as the iPod) don't play Ogg Vorbis. Grrrrrrr. --Cyde 09:17, 7 November 2005 (UTC)
  • After a while, the articles at the end of the list will need to be re-recorded to reflect changes to the text version. A date-ordered list would make it clear which articles likely need re-recording.

Does anyone have ideas about how to implement this? -SCEhardt 22:56, 24 September 2005 (UTC)

How can I help?

Previously I've nominated an article (Sun Yat-sen) into FA and it's going to be on the main page 8 days later. However, up till the moment, there was no spoken version for this article and I'd like to record it myself. How can I help? Deryck C. - the very original one

Deryck C. - the esperanza-enriched one 10:08, 30 September 2005 (UTC)

Template for spoken version of main article

I have adapted the existing template for the top of articles into one that can be used to complement the template main. To use it, write:

Listen to the main article: Doctor Who theme music · (info)
Spoken Wikipedia
This audio file was created from an article revision dated 2005-10-07, and does not reflect subsequent edits to the article. (Audio help)
{{Spoken Wikipedia Main|FileName.ogg|YYYY-MM-DD|Main article name}}

It is currently in use on the main Doctor Who page. If no-one objects, I shall add this template onto the main Spoken Wikipedia project page in 7 days. throup ✆ ✍ 20:04, 7 October 2005 (UTC)

I have now added this template onto the main page. throup ✆ ✍ 14:07, 15 October 2005 (UTC)

Concerned about my age

I would really like to do Columbine high school massacre, it being an event which led to zero tolerance which I am strongly against. However, I am concerned about how it would be taken if someone of my age would be doing a recording for an article that serious. Syckls 03:55, 9 October 2005 (UTC)

I am not an authority on this, but I would say go for it. If you're really not sure about it, post a sample here before adding it to the main article. There is no reason why someone your age (14) cannot speak about a “serious” issue. throup ✆ ✍ 07:26, 9 October 2005 (UTC)
Shoot, now I realize I even forgot to put in my name the second time..... Anyway, thanks. I'll claim it on the project page. Syckls 01:27, 10 October 2005 (UTC)

On the nature of organic articles

I fail to see how spoken word recitations of articles will be a practical feature. While the viewing-impaired users will appreciate said function, what is to happen to spoken articles as the articles are edited? --Unsigned by 68.104.11.5 01:28, 16 October 2005 (UTC)

If an article is in constant flux, it's probably not a good choice for reading. However, when a spoken article is tagged on a page it is dated and can also be linked to the revision it was taken from. This gives us two advantages:
  1. It is clear whether the recording reflects the current status of the article.
  2. Even if it doesn't reflect the current status, as long as a quality article was chosen in the first place then we still have a quality spoken article.
It does raise the question, though, should we have a section on the project page for articles in need of updating in order to best match their written form? --throup ✆ ✍ 08:51, 16 October 2005 (UTC)
We can just use the request page for this. If we find an article that has changed enough that a new spoken version is in order, we can put it on the requests page with the reason: 'new version needed'. CB Droege 14:23, 16 October 2005 (UTC)
I would appreciate anything that makes it easier to keep track of the progress of the articles that I've done spoken versions of. --Macropode 05:00, 13 November 2005 (UTC)

Playlists

What do people think about using playlists when a spoken article is split into multiple files? For example, I have recently recorded a spoken version of Dalek, which I ultimately split into three parts. Obviously the recording is split for convenience due to issues of file size and also recording duration, but I also think it would be convenient to link the parts together.

It is easy enough to knock up a M3U playlist (M3U appears to be the most suitable as it is an ad-hoc open format and appears to have the most support from applications) but I can't find an easy way to deliver such files. The best I could manage was the following:

Dalek/m3u (the playlist source--to view this "unwikified", go to edit)
Dalek.m3u (the playlist itself)

In some (technically inferior) browsers this will be identified as a file named Dalek.m3u and hopefully dealt with accordingly. However, most browsers will see a plain text file named index.php and have no idea how to deal with it. If this happens, as is likely, save it as Dalek.m3u and then load this playlist into any compatible media player.

I would be interested in others' thoughts on this and also any suggestions on how to improve this delivery method. --throup ✆ ✍ 23:21, 25 October 2005 (UTC)

To work properly from the mediawiki servers, apache would have to be configured to deliver m3u files with the correct content-type. That's elementary (assuming we can persuade the devs to do so, it's just a single text file) - but we need to find out what the correct type to send is. Right now they're serving "text/html", so it's not unreasonable for browsers to not do anything meaningful with it. -- Finlay McWalter | Talk 23:40, 25 October 2005 (UTC)
The standard content types for M3U files are audio/x-mpegurl and audio/mpegurl. --throup ✆ ✍ 23:50, 25 October 2005 (UTC)

Choice of Audio Codec

Why is Ogg Vorbis the preferred format? iPod, the dominant player in the field, can't support it, and we're missing out on a huge podcasting segment of iPod people who might want to download and listen to Wikipedia articles but can't. Is the whole Ogg Vorbis thing a requirement or a suggestion? Becuase if it's not a requirement, I'm going to go for mp3. And I understand that you guys want to be free and everything and embrace totally open source technologies ... but I think that, in this case, mp3 is still the way to go. --Cyde 09:14, 7 November 2005 (UTC)

I support this. You shouldn't put your content in a format the large majority of audio players can't support. JG of Borg 00:02, 13 November 2005 (UTC)
Ogg Vorbis is, presumably, preferred because it is an open, patent-free format (and so, theoretically, should be universal), and generally produces smaller files than MP3. In practice, however, MP3 is the standard compressed audio format on today's computers and portable devices. Therefore, I support changing the standard format of this project to MP3. --LostLeviathan 22:09, 4 December 2005 (UTC)
I don't think MP3 is compatable with Wikipedia for legal reasons. See the licensing and patent issues mentioned here. -SCEhardT 23:32, 4 December 2005 (UTC)
MP3 is out of the question because it requires patent royalties to encode and decode. I don't deny that MP3 is by far the most widely used format, but that doesn't make it "the way to go"; and surely whether some particular esoteric device supports it or not is wholly irrelevant. — I strongly urge everyone to use Ogg, and I strongly urge everyone who spots someone else uploading an MP3 to download it, convert it to Ogg, and delete the MP3. — Timwi 16:00, 5 December 2005 (UTC)
I fail to understand your objection to the MP3 format. Wikipedia itself does not have to encode or decode the files, and while the developers of encoding and decoding software must pay royalties, in practice, the vast majority of computer users have the capacity to do so, while fewer can use Ogg. Examples (feel free to add information to each of these individually):
  • Microsoft Windows, the dominant operating system, plays MP3s out-of-the-box via Windows Media Player. MP3 Encoding can be done with iTunes, a free program. Additional software [1] must be installed in order to play or encode Ogg Vorbis files.
  • Mac OS plays (and encodes) MP3s out-of-the-box via iTunes. Additional software [2] must be installed in order to play or encode Ogg Vorbis files.
  • Linux -- not sure.
  • Palm (PDA) - Palm users are now able to surf the web with their devices, and can thereby access Wikipedia. MP3 play is supported out-of-the-box. Ogg Vorbis requires additional software.
  • Pocket PC - Similar to the situation with Palm devices. Many can access Wikipedia and play MP3s out-of-the-box, whereas none (to my knowledge) include software to play Ogg Vorbis.
  • iPod and other MP3 players - All support MP3 files. A minority support Ogg Vorbis files. This is not "wholly irrelevant"; this is the difference between many people being able to listen to Wikipedia articles while commuting, or only being able to listen while they're on a computer.

If there is a legitimate argument against MP3s, could you please make it in a practical form, rather than expressing it theoretically. --LostLeviathan 03:46, 6 December 2005 (UTC)

It's policy. See this, linked from Wikipedia:Media#Audio. - mako 01:18, 12 December 2005 (UTC)


Linux actually prefers OGG over mp3 (for patent/rights issues, like Wikipedia). Some popular distributions, like Ubuntu don't even support MP3 out of the box. People are having similar discussions on their website. Ckamaeleon 20:20, 19 January 2006 (UTC)


From my personal POV the perfect codec for the Spoken Wikipedia Project is Ogg Speex rather than Ogg Vorbis. The reasons are mainly 3:

  • First of all both the encoder and the decoder are free software, so the policy is safe.
  • Second: Speex is one of the very few open source audio codecs specifically designed for speech compression, resulting in very clean recordings with a fifth of the file-size compared to other general purpose algorithms, like Vorbis.
  • Third: the portability is clearly not an issue, so why not to use the perfect match with your needs, if there's one? Speex encoders are available for every platform (as for Vorbis, of course), Speex plugins have been released for many (not saying all) software players.

Given that Vorbis is a niche, Speex is even a smaller niche, but apart from that, I see no reason for preferring Vorbis over Speex as long as the project is targeted upon spoken content. --Nosferatu it 23:29, 26 December 2005 (UTC)


It may be worth observing that there is an increasingly large community of small computer users for whom Free/Open Source software (most commonly in the form of Linux and friends) provides the only affordable and logical alternative to the dominant licensed proprietary software model. From a personal point of view, if it weren't for Free standards such as Ogg Vorbis (or, for that matter, Speex) which by virtue of their patent-free status are able to be legally implemented and used freely here in Australia (unlike MP3), it would simply not be possible for me to contribute to the Wikipedia, amongst other things. --Macropode 04:49, 30 December 2005 (UTC)

Can everybody do Spoken Wikipedia?

Just saying, can everybody help on Spoken Wikipedia, no matter the age, accent etc? I am at a young age and would love to help. Uncke Herb 05:31, 8 November 2005 (UTC)

I'd say anyone is welcome to contribute as long as they can make a clear recording and are willing to take the time to edit it properly. -SCEhardt 22:14, 8 November 2005 (UTC)
I'd second that. throup ✆ ✍ 22:39, 8 November 2005 (UTC)
Perhaps this should be on the front page to encourage people who are hesitant on recording a wiki article? Gflores 23:22, 8 November 2005 (UTC)
Of course, not everybody. For example, I myself who doesn't have the money to buy a high-class microphone cannot. Deryck C. 08:07, 9 November 2005 (UTC)
I don't think it takes a high-class microphone to contribute. If you take a look at the sample configurations, many people are using pretty basic microphones. I'm not sure what is available in Hong Kong, but I was able to pick up a decent computer mic for only $14 in the US. -SCEhardt 16:15, 9 November 2005 (UTC)
My mic is only $18 HKD. Deryck C. 02:29, 11 November 2005 (UTC)
I agree with SCEhardt, as long as it's understandable and well recorded. The world is nice because it's different, afterall.Federico Pistono ✆ ✍ 05:16, 11 November 2005 (UTC)
I'd have thought that you might as well use whatever crappy mic you can get your hands on. If it doesn't work, then go out and get a different one, but you never know until you try... Odd bloke 03:53, 14 November 2005 (UTC)
For curiosity, my mic's only AU$5. It's still pretty good. Uncke Herb 08:26, 21 November 2005 (UTC)

Wav files?

What about those of us who use someone else's computer and hence can't use specialist software. When I was able to use my home computer, I used winamp etc. to make recordings, now I use my office computer so I am inevitably stuck with Microsoft software, which inevitably gives me the .wav format to use. So can we only use the .ogg format or would .wavs be acceptable? DavidFarmbrough 13:28, 22 November 2005 (UTC)

Wav files are huge! I'd say no because of that if nothing else. However, if you (or anyone else for that matter) need a wav file converted to ogg, let me know and I'll be happy to do it for you -SCEhardT 14:48, 22 November 2005 (UTC)
Better yet, instead of requiring people to let you know, put Category:Spoken articles on your watchlist and whenever someone adds one, check if it's a WAV file and if so, convert it. — Timwi 17:47, 22 November 2005 (UTC)
I'm certainly in no position, nor do I have any intent, of requiring anyone to do anything :-) The problem I see with uploading wav files is that they are so large (about 10 MB/minute). I think such large files would exceed the upload size limit for Wikipedia or the Commons, so I am offering to provide a temporary login to a webserver that could hold such a file. However, it would certainly be simpler to go ahead and upload the wav file to the Commons and convert it later, so if that works I say go for it! -SCEhardT 20:03, 22 November 2005 (UTC)
Wav files are huge if you use the highest resolution stereo version, with something like spoken word articles this isn't necessary. An 8 bit mono is fine for speech. BUt I do appreciate the offer, SCEhardt, thanks. DavidFarmbrough 10:12, 23 November 2005 (UTC)
I would suggest the solution of storing your WAV files on the internet somewhere, then converting them to Ogg Vorbis or MP3 files (see discussion above) from another computer. While WAV files are somewhat universal, they are just too big. --LostLeviathan 18:02, 7 December 2005 (UTC)

ogg file summary template glitch?

Help! I just submitted my first contribution to the WSW, by uploading it to the Commons using the template suggested on this page. Everything seems to have gone smoothly, except that where there should be my user name, instead {{{user_name_link}}} is showing up. Can anyone help me fix this? The file is at http://commons.wikimedia.org/wiki/Image:En-Rosa_Parks.ogg  :: Salvo (talk) 08:39, 29 November 2005 (UTC)

I think I have the username fixed now - I set the link to your Wikipedia page but if you want the Commons user page instead, just take out the w:. Also, I changed the article link so that it points to the Wikipedia page, but I can't figure out how to keep it from displaying w:Article instead of just Article. -SCEhardT 17:16, 29 November 2005 (UTC)
Thanks for that, I guess I could have put in a link like that myself, I just figured the template was somehow automatically formatting stuff, or something. Or maybe not. I'm not always the best at this coding stuff. :)  :: Salvo (talk) 04:00, 30 November 2005 (UTC)

Columbine Massacre

Hello, I placed a request for this article on September 22 and I wanted to know what the progress for it is. I am not sure if this is the correct place to ask this. Any information would be appreciated. Thank You. -- PRueda29 Ptalk29 04:46, 04 December 2005 (UTC)

From above, it looks like Syckls was planning to work on this article, so you might want to contact him about the progress. However, it looks like Syckls has been away since October 30th. If you can't get in contact with him, you might want to strike the article name and make a note at the active participants list so that other contributors don't think someone is working on the article. -SCEhardT 05:15, 4 December 2005 (UTC)
Well, anyone willing to take the Columbine High School massacre would be appreciated! -- PRueda29 Ptalk29 05:31, 15 December 2005 (UTC)

Not everybody should be doing spoken articles!

Spoken articles should only be used for disabled individuals such as blind people etc... However I would really, and I mean really reconsider the people who are chosen to record and used to read out the articles, because so far they are all below average, with slurring, mis-reads and weird accents. You can't expect just anyone to do these with their computer mics (resulting in reverb and irritating interference sounding like udnerwater star wars music), but I would strongly advise using professionals with sound-design and recording experience or audio-engineering instead. Because as of now the majority of the recordings I have heard are very poor indeed! Piecraft 00:58, 16 December 2005 (UTC)

You should point out the issues, e.g. on the talk page of the recording, so that the contributors can try to fix those problems. Mis-reads should have been cut out of the recording and re-read, and the contributor may have the original file and be able to fix the problem easily.
We could perhaps implement some system to rate the technical and reading quality of recordings so that those that need to can be re-recorded (by another person if neccesary).
Unless you're paying, or they're giving their time for free, there'll be no professional recordings. There is no reason for accents to be an issue though, unless they're incomprehensible. Indeed, IMO, accents should be encouraged for entries that are relevant to specific geographical areas. Joe D (t) 01:24, 16 December 2005 (UTC)
I consider the accents rather nice. Even if a particular topic is not region-specific, having different accents makes the recordings more interesting. ~MDD4696 01:20, 6 January 2006 (UTC)

Spoken articles should only be used for disabled individuals - Why is this? I really enjoy listening to the spoken articles while walking to work and think that they are for everyone to enjoy. The Spoken Wikipedia podcast is out there for all podcast lovers to hear; surely it is not for only disabled people. -William Morgan

Special recording and playing software for encyclopedia contents

1. Marking paragraphs, or even sentences if the lector wishes.
2. Marking links.
3. Automatic checking for updated parts of the text, and marking it for possible

re-record.

All this supported also by the playing software. Even the listener might tag possible errors in what he or she listens.

Some portable technology working within a browser, like Java.

What do you think?

Hello! Is anybody here? ;)

I might try do write a very simple open source app with a subset of these proposals, just to test it, but I'd like to know if anybode here would possibly be interested.

Image:Bulbasaur.ogg

Okay, I just recorded Bulbasaur, what do I do with it now? --Celestianpower háblame 16:21, 25 December 2005 (UTC)

Recordings should link to the version they're based off of

Personally, I think this should go without saying, so as soon as I get a better handle with the templates assorted with this WikiProject, I'm going to be bold, but here's what I think:

There is some discussion about what should be done when a spoken version of the article becomes outdated, and some of the templates mention it. However, some entries unsatisfactorily list only the date of the revision that was recorded. While it is possible to trace the correct revision with some sleuthing, it would be far more helpful if the actual link to the revision of the article the recording was based on was provided. Some templates indeed do it, but coverage is spotty. So we should make this much clearer.

I believe this would entail adding another entry/more documentation to the templates: {{Spoken article entry}}, {{Spoken Wikipedia}}, {{Spoken Wikipedia-n}}, {{Spoken Wikipedia Main}} — Ambush Commander(Talk) 00:37, 29 December 2005 (UTC)

Where are all you women?

It's getting a bit blokey (male-dominated) here in Spoken Wikipedia. :) --Macropode 03:47, 30 December 2005 (UTC)

Speech sytnthesis

What's about a speech synthesis solution, like (an opensource solution like) Readspeaker, as it is used on the site Agoravox (in french) : the synthesis is done server sidde and the resulting mp3 transmitted to the client. So, it's probably to complicated/expensive : the hardware infrastructure to make that possible should be pretty heavy... Traroth 09:32, 4 January 2006 (UTC)

One of the reasons we decided to start the project is because many people found listening to synthesised speach painfully tedious! Joe D (t) 23:07, 8 January 2006 (UTC)

Does anyone know how to convert a wave to OGG? second, could we start an article on that? --CyclePat 03:36, 9 January 2006 (UTC)

http://www.vorbis.com/software/ Joe D (t) 03:51, 9 January 2006 (UTC)

Seems to me that the article Speech synthesis could use a version of the article read using a good speech synthesizer. --Carnildo 02:50, 11 January 2006 (UTC)

That actually tickles my sense of humor :D I have TextAloud with the Natural Voices installed - may I should run the article though it ;) - Beowulf314159 02:55, 11 January 2006 (UTC)