Wikipedia:WikiProject Spoken Wikipedia/Pronunciation task force

Welcome to the Pronunciation task force of WikiProject Spoken Wikipedia.

Scope

 * The pronunciation task force aims to supply short recordings of difficult-to-pronounce terms. These recordings will supplement the IPA transcriptions recommended by Manual of Style (pronunciation) and described in Help:IPA/English.

Participants
Please remember to also list the languages you speak.

Active

 * : French (Paris, media)
 * : English, French (New Eng. Fr.)
 * : Russian
 * : English (California accent), specialty: medical terminology
 * : En (General), Np (native), Hi, (Devanagari Texts, Little knowledge in Spanish)
 * : Swedish, English, German
 * : English, Hindi, Urdu
 * : English, German, French, Italian
 * English, Hindi
 * : English, German, French, Italian
 * English, Hindi

Inactive

 * : English, Tamil
 * : English, Tamil
 * : English, Tamil
 * : English, Tamil
 * : English, Tamil
 * : English, Tamil

Making sure existing pronunciations are in use
Browse through some of the following places to look for recordings that we might be able to use on the English Wikipedia if we're not already. If you find one of use, add it to the article along with its IPA transcription as described under Add recording to article with IPA below. This is a great way to contribute if you're not able to do recordings.


 * wikt:Special:WhatLinksHere/Template:Audio, wikt:Special:WhatLinksHere/Template:Audio-IPA: Articles on the English Wiktionary including pronunciations.
 * Commons:Category:Pronunciation, particularly Commons:Category:English_pronunciation
 * Special:WhatLinksHere/Template:wiktionary: Articles using the template wiktionary may link to an entry including a pronunciation.

Finding an article to contribute a pronunciation to
Some places to look:
 * WhatLinksHere links for the various general IPA templates: IPA, IPA-en, pron-en, IPAlink-en, IPA-all, pronounced, IPAlink, IPA2, IPAEng, pronEng
 * WhatLinksHere links for the language specific IPA templates: Australian English, Australian English 2, Hebrew, Irish, Korean, Polish, Russian, Spanish
 * wikt:Category:Requests for audio pronunciation
 * wikt:Wiktionary:Pronunciation file requests
 * Categories of people, cities, and other topics with proper names, particularly in other countries.

Check for an entry on the term in the English Wiktionary and its native language Wiktionary, if applicable, to see if it already has an audio pronunciation and/or IPA pronunciation listed. If it has an audio pronunciation, just use that and skip to Add recording to article with IPA below (unless you wish to improve upon it).

If you find an article that needs a pronunciation but you don't feel qualified to add one, use Pronunciation requested audio.

Record pronunciation
If you're not sure how to pronounce the term, try Merriam-Webster or howjsay.com for an example (but of course do not copy IPA or sound files directly from non-free websites). For many terms, you may be able to find videos online where people pronounce the name correctly (but be wary of incorrect pronunciations). In some cases, such as names of people, you can contact the person directly and ask them, or even ask them to record it and donate it themselves via OTRS.

Record a pronunciation in OGG format. Much of the advice at WikiProject Spoken Wikipedia/Recording guidelines applies here (e.g. equalization and noise reduction), except that unlike a spoken article, a pronunciation recording should contain only the pronunciation of the word, and no English description or explanation. This allows it to be reused across projects. The pronunciation should begin with half a second of silence, and finish with one second of silence, to avoid truncation by some players. See also wikt:Help:Audio pronunciations.

Upload the pronunciation to Wikimedia Commons using the Upload Wizard. At the "Release rights" step, it is recommended to select "Use a different license" and then "Creative Commons CC0 Waiver" &mdash; because audio pronunciations are very short, the requirements imposed by other licenses can be problematic. At the "Describe" step, for "Title", use the format "language code-dialect code-term", for example "en-us-Euphrates". In the description, include the term being pronounced, a description of your speech dialect, and any sources you used to determine the correct pronunciation, if applicable. Add the appropriate subcategory of commons:Category:Pronunciation based on the language of the pronunciation (e.g. ).

Add recording to article with IPA
If there is not already an IPA transcription, and the term is English-language, we strongly recommend using IPAc-en, which both allows audio to be included, and also allows the reader to hover over each IPA symbol to give more information, as in these examples from Template:IPAc-en:



The individual IPA segments like "æ" are taken from the tables at Help:IPA/English. The template should be added in parentheses directly after the article title in the lede, or if there are already parentheses after the lede, insert the pronunciation as in this example:


 * Marc Ambinder (born c. 1978) is an American editor and journalist ...

For non-English terms, it is generally necessary to use a language-specific template, which does not provide hover tips. For example:


 * → Swedish: åka,, 'travel, go'

Finally, visit the file description page for your uploaded .ogg file, and edit it and copy your IPA transcription into the description there as well.

Add recording to other projects

 * Check the English Wiktionary to see if it has a listing for the term; if so, add the IPA transcription and audio recording there as well using the and  templates as described at wikt:Wiktionary:Pronunciation.
 * Check the interwiki links on the Wikipedia and Wiktionary entry, particularly those corresponding to the "native" language of a foreign term. If you don't know the templates to use on a particular non-English wiki, just use the following markup: (IPA transcription here)

Categories

 * Category:Pages including recorded pronunciations
 * Category:Wikipedia requested audio of pronunciations