Wikipedia talk:WikiProject Spoken Wikipedia/Recording guidelines/Archive 1

Volum Levels covered twice

Volume Levels are covered in the Production tips section. I like the lower section because it sounds more basic and accessible, but the subject of sound levels has already been covered, so this is redundant. The top one is also more professional sonding and concise. the main differance between these two is that the bottom one talks about levels before/durring recording while the top one is for editing. These should be merged some how. Perhaps the bottom one should be severly shortend and called, "volume levels durring recording". on the other hand, the top one could be pulled out of its section and put into the sound levels section.

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 whol 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 08:17, 7 November 2005 (UTC)

quoting from the Media page, "Wikipedia uses Ogg Vorbis for sound. The Ogg Vorbis format is not encumbered by patents, and has been found to offer higher audio fidelity than the MP3 format. It is competitive even at low bitrates for encoding speech, comparable with even HE-AAC. A decision has been made that MP3 files will not be hosted at Wikipedia."

While you have an excellent point about iPod listeners + OGG, I think it might boil down to legal issues at the end of the game. There's probably page where people are discussing this more extensively on Wikipedia. Personally, I'd prefer MP3 as well, but now that I'm running Linux in addition to Windows, it's not too difficult to work in OGG. Ckamaeleon 19:57, 19 January 2006 (UTC) (sorry about leaving out the "not" earlier. I only just now caught it.) Ckamaeleon ((T)) 15:21, 24 April 2006 (UTC)

I'd be happy to donate my time but it's just way too complicated. I'm still not sure how I'd upload a file even if I knew where to enter the script to reference to it in an article. You guys must be able to come up with a form linked from an article page that would allow me to enter all the required information including the file location on my computer for upload and then just click an Upload button to be done with it.

on Relaxation

Stay relaxed. Whenever you realize that your voice is becoming tenser than normal, or you're having more and more trouble speaking fluently, take a break, and resume when you feel muscular tension going away. Keep some fresh (not cold) water into reach, and drink some whenever you feel your mouth drying out.

Is this not the most patronising paragraph you've read in a while? Obviously the kind of people with the dedication and technical know-how to do this are going to be capable of understanding when they need to relax...

What technical know-how? Plugging a mic in and following instructions in Audacity is hardly splitting the atom. I think it's important to encourage people to do this as far as possible. Odd bloke 04:00, 14 November 2005 (UTC)

No, I think this is good advice. It's not meant to be patronizing. Vocal performance instructors will tell you the same thing. People today aren't conscious of their bodies when they're speaking, even though muscle tension makes a difference in their sound. Sometimes, it's good to be reminded of the basics. Ckamaeleon 19:29, 19 January 2006 (UTC)

What to Read

So, obviously an article in Wikipedia contains a lot of stuff besides plain text. We've already covered that one should read the things in parentheses, but I think it would be helpful to add more information about what to read on a page.

For instance, how does one approach:

• references to visuals/captions
• tables
• subsections... "subsection 1, Variations" vs. "section 3.1, Variations" vs. "...Variations..."
• equations
• The entire references section of most articles
• InfoBoxes
• ...

If there's stuff already written about this, my apologies. I didn't find it when I used the Beginner's guide. I also think a tutorial on recording would be grrrrreat. Ckamaeleon 19:36, 19 January 2006 (UTC)

I agree, this needs to be clarified... in this request - Talk:Zero-point_energy - the user asks for the reader to add his/her own contributions to clarify meaning. Am I right in thinking that that shouldn't happen at all? It should be a word for word rendition of the article. --155.198.36.23 13:57, 8 March 2006 (UTC)

You're right, that definitely shouldn't happen. That request looks like it would be better made as a request for the article in general. If it's too technical, then the article itself should be cleaned up and clarified, and then a recording made from that. -- Laura S | talk to me 14:55, 3 August 2006 (UTC)

Infoboxes are the most frustrating, and it's preventing me from reading some articles because I don't want to leave them out, but reading them would be so tedious. I think in the zero-point energy article the person was probably referring to the equations, which most non-physicists would be unable to read. --Omaryak 22:00, 2 August 2006 (UTC)

I'm not sure if there's any great way to handle that sort of thing. I've generally omitted infoboxes and other tables full of information. It's hard to read them in a way that makes them useful to the listener, and especially long tables would get boring. Equations are also rough and I've omitted those, partly because I'm not sure of the correct way to read them, and partly because they're so visual that a listener would likely not find them helpful. For astronomy articles that's not such a big deal, but might become an issue in a mathematics article that depends heavily on equations. Lastly, whenever a sentence references an image (e.g., "The picture to the left shows..." or "As seen in this photo"), I leave out the reference altogether to avoid confusing the listener. -- Laura S | talk to me 14:53, 3 August 2006 (UTC)

Windows Microphone Boost

I found that the audio quality is much better without using Windows Microphone Boost. My first recording was done with the boost on and not much amplification in editing, when the second one was without the boost and a bit more amplification. It's so clearly better that I almost feel like recording the first one again! It's in Volume / Options / Properties / Recording / OK / Advanced Control / Advanced / Microphone Boost.

OGG settings - link to where?

This page says, We recommend that you use the following settings: 48 kbit/s, 44.1 kHz mono. (see talk for how to achieve this). Talk links to Wikipedia_talk:WikiProject_Spoken_Wikipedia. I just read three archives and the current talk and I'm still in the dark, so please make this instruction more specific, it's very frustrating. >:( I suspect it's referring to Archive 1 but there doesn't seem to be a good guide esp. for setting the 47 kbit/s bit. Thanks, pfctdayelise (translate?) 07:08, 14 April 2006 (UTC)

Which recording program are you using? -SCEhardT 12:40, 14 April 2006 (UTC)

Only the one that everyone else here seems to be using...Audacity. pfctdayelise (translate?) 12:01, 18 April 2006 (UTC)

Hrm - come to think of it I'm not sure how to get those exact settings in Audacity. You can set the 44.1 kHz by going to File-> Preferences-> Quality and adjusting the 'default sample rate'. You can also change the output file size by using the slider under the 'file formats' tab to set a higher or lower quality. However, as far as I know, ogg changes its kbit/s rate based on the complexity of the audio (so you get a consistent quality, not consistent kbit/s). Hope this helps! -SCEhardT 21:55, 18 April 2006 (UTC)

I just used Audacity for the first time and it used those settings by default. Aguerriero (talk) 22:15, 6 June 2006 (UTC)

File names for multiple parts

So happy to have done my first spoken article! It says the filename has to be the name of the article plus .ogg, which makes total sense. However, for multiple parts, how do you name them? I named mine the article + Part_1.ogg, etc. This works for linking from the article, but in the info page doesn't work, because it's looking for only one file. How do you accomplish multiple parts without messing up the filenames? Doing it with all one file made it so big it timed out on the upload, plus the instructions suggest splitting up larger files anyway. Thanks! --Laura S 00:09, 30 April 2006 (UTC)

As far as I know the talk page template only accepts one file - maybe a template for multi-part spoken articles would be a good project for someone? -SCEhardT 00:45, 30 April 2006 (UTC)

I'd like to strongly advocate for changing the template in some clever way to allow for sectional recordings. It will make the recording task much more practical, and is a great way for the audio wiki to act more like a text wiki and make it easy to change something someone else did. Right now if I change one word, then the whole audio file is out of date. But if I were to change a paragraph, which was, say, half of a section, then someone *else* could re-record the new section and leave the others intact. —The preceding unsigned comment was added by Vonfraginoff (talk • contribs) .

Could you elaborate on this a little? It sounds like you're advocating recording articles paragraph by paragraph or in similarly small chunks. This would make later updating simpler, but it has some drawbacks. Primarily that recording it in the first place like that would be significantly more difficult, time-consuming, and error-prone. It would also result in a recording that, when pieced together, lacks (for lack of a better term) vocal flow. It might end up sounding less like one coherent pass of the text and more like several different paragraphs jumping around.

My other concern is that even if the initial article turned out beautifully, once others come in to edit bits and pieces of it, it will start to sound like a patchwork of voices. It would be very distracting to a listener to have completely different voices, qualities of recording, volumes, etc. in different sections.

The lack of editability is definitely an issue, especially for less stable articles. However, I think having different people come in and change parts of a recording, while wiki-like, would result in a poor quality recording. I've been keeping the raw files of my recordings in case parts need to change, but really it would be best just to re-record the whole thing once a recording becomes too out of date. -- Laura S | talk to me 14:06, 17 August 2006 (UTC)

For the sake of listener's ears there should be only one narrator per article, the one who has the original, un-compressed recording. However, there isn't any lack of editability in spoken articles, just lack of experience in using your chosen audio editing software. When you're familiar with your tools, you can concentrate on the art rather than the process by which you produce that art.

If you're a naturally good narrator, recording the entire article in one or two takes can produce great results. The rest of us can resort to recording small sections, possibly editing out the flubs as we go, and joining the sections together. You can get great results this way too at the cost of having to learn a few more editing tricks, but this knowledge pays off when modifying your recording later to keep it current with factual changes in the text article.

Editing bits out and putting new bits in to an existing recording is easy and fun, once you've climbed the learning curve. Sermon ends. -- Macropode 03:08, 20 August 2006 (UTC)

Shorter articles...

Would it be possible to get a list of shorter articles to record for newbies? The thought of doing a FAC article for my first one seems a bit intimidating. I did go look at the Articles All should have list, but even there, it's hard to figure out which ones I really ought to look at. Thanks. Wikibofh_(talk) 20:47, 16 June 2006 (UTC)

My suggestion would be to start with an article that interests you. The only real caveat is that you should try to select an article that's pretty stable; otherwise, the text may change significantly and soon, leaving your recording obsolete. -- Laura S | talk to me 01:05, 17 June 2006 (UTC)

• • Thanks, I've been working on one, but still learning. My point was that if there were a list of shorter articles it might be easier for us spoken newbies. After all, it's not like I'm a wiki-babe-in-the-woods.  :) Are there any particularly good examples I should be paying attention to? My only example so far is Helen Keller. Wikibofh_(talk) 02:20, 17 June 2006 (UTC)

Software?

Can anyone recommend some good (free) software for doing the recordings on a Mac? I used Audacity on my PC and loved it. I've switched to a Mac (Tiger), and Audacity runs, but is very flaky and crashes nearly every time I run it. The other tools suggested are fairly expensive. A quick Google search shows there are plenty of tools out there, but I know nothing about them; can anyone recommend one? I'm going to keep trying with Audacity for now but I'm really worried I'll be twenty minutes into a recording and it will crash when I try to save or something! -- Laura S | talk to me 01:57, 19 June 2006 (UTC)

Isn't there a native application that comes with Tiger that does that stuff? Maybe I'm mistaken. Anyway, it looks like Audacity is the best free application.. which of course, isn't working very well for you! Aguerriero (talk) 15:36, 19 June 2006 (UTC)

Hmmm, I'll have to check it out and see what there is. Maybe that's what Garage Band is for? I'm still figuring out what's on there :) Thanks! -- Laura S | talk to me 17:02, 19 June 2006 (UTC)

I have a special application for my Mac called Orator. It allows you to insert text and then press Speak to hear it spoken (with a Mac OS X voice, of course) or export it to an AIF file (which I can convert to Ogg.) The quality of this text-to-speech rates about 4.5 of 5. Could I use this to produce spoken articles quickly?

GNU

Instead of releasing our sound files under GNU, can we relase them into the publc domain? American Patriot 1776 23:04, 19 October 2006 (UTC)

Nope - since the text is licensed GFDL, all derivative works must also be GFDL. -SCEhardT 02:42, 20 October 2006 (UTC)

Choice of Audio Codec: Speex?

What do you think of Ogg Speex? Wouldn't that codec be a good choice for the spoken articles?--Imz 15:45, 19 May 2007 (UTC)