乡音苑 Phonemica
 
sign in
 
UI language:
Select language family:
A panorama of local languages, through stories told by their speakers
 
Mandarin : 171
Wú (Shanghai) : 57
Cantonese : 36
Mǐn : 35
Hakka : 23
Gàn : 23
Jìn : 21
Xiāng (Changsha) : 16
Píng Hé Tǔ Huà : 2
Hui : 1
Multiple Dialects : 1

recent blog posts

Announcing Phonemica 2.0

Today marks exactly one year since our official launch. In that year we’ve seen around 400 recordings come in, we’ve had over 9000 users register. We’ve been fortunate enough to meet many of our users and volunteers, and we’ve had a hell of a lot of fun doing it. With that in mind, it’s time for an update. Because the other thing this past year has shown us is how to improve the project. And so, with that in mind, we’ve been working our butts off these past two weeks to completely rewrite the site from the ground up. The database is new. The code is new. And hopefully you’ll notice many of the improvements we’ve been working on. But in case any of them aren’t obvious, we’d like to take a moment to introduce to you some of the changes.

What’s New in the Update?

The Map

  • Brand new map system, free of the Google API that was causing some problems for users in China
  • An all-new map design just for Phonemica
  • You can now search the map by location. Jump straight to your hometown and see what’s nearby
  • Faster map filtering
  • Use the map to search recordings for keywords. Want to see how many of our speakers say the word “fishmonger”? Now you can, plotted right there on the map.

Dialect classification

The site previously was following the Chinese Academy of Social Science’s 1987 classification. We’ve always known we wanted to update that, and have met with members of the Academy numerous times in cooperation to this end. However after a few snags and delays, we’re finally going ahead with an updated classification. You’ll see this reflected in some dialect name changes, and a few reassignments of recordings.

Backend upgrades

  • much more responsive transcript scrolling and timing
  • database queries have been optimised, meaning less chance of the site being overloaded in peak times
  • rewritten site code means more stability and less chance of stray bugs left over from very early versions of the site that were never really sorted out until now.
  • Uploading

If you’ve submitted a recording in the last couple months, you’ve already seen an improvement in the uploading process. We’re not done improving it. We’ve taken it a step further, making the whole process simpler and faster for you, with more improvements being planned after some initial testing.

Managing your own recordings

You now have the ability to edit titles and subtitles of your own recordings. We’ll soon be adding the ability to do more recording management for those that you’ve uploaded, which I’ll list below.

Additional site localisations

Before, if you wanted to view the site, you were limited to English or Chinese, in either simplified or traditional characters. We’ve added the ability to add further localisations, and we’re already mostly done with a Korean translation of the interface. This will allow a much larger user base to view the site. If you’re interested in helping us create further localisations, please get in touch with me at keyilan@phonemica.net and I’ll be happy to work with you to set up the interface in other languages.

Coming soon

You didn’t think we were done there, did you? We’ve actually got a lot more that we’ve been working on recently and that we hope to make available to you in the coming month. However in order to make it as easy to use and as useful as possible, we’re still doing a bit of testing, so we’re not releasing it just yet. But I can still tell you about it.

Re-segmenting your recordings

Coming very soon is the ability to edit the timing of segments on your recordings. In other words, you can adjust the timing of the phrases in the transcript, making the segments line up better with the actual words and phrases of the speaker. This has been something we’ve wanted to improve for a long time, and in doing so to give you more control over your own submissions. Here’s a quick sneak peek at the tool:

Streamlining the upload process

Submitting recordings is still not as clear and easy as we want it to be. We’re still working to improve the process to make it totally straightforward. We know any extra time you have to spend on the process is that much more to deter you from actually taking that time, and so we want to remove all the obstacles. Already it’s simpler than it was a week ago. We’ll be improving it even more in the coming month.

User-based mapping updates

We’re also working on a system to allow users to make changes to the map data. Since much of the base data has come from open source data sets, there are sure to be some mistakes or misplaced towns. In the coming weeks we’ll be opening up the ability to adjust locations to users so that you don’t have to worry about your town showing up in the wrong place.

The biggest change of all

Finally, there’s one more really big thing that we’ve changed. Many many people have asked for this, but due to system architecture issues we could never quite do it with the old system. A big motivation for the re-write of all the code was to make this possible.

Today, we’re officially announcing a major change to the site’s focus:

Phonemica is now no longer just for Sinitic dialects. That means your Zhuang and Hmong and Korean recordings now have a home.

We’re starting small, since each new language to the site still requires a fair amount of planning and infrastructure, not to mention extra coding for things like transcription and classification, but we’ve been working hard to make sure everything is in order. We’re most of the way done with the functionality for recordings in Korean, and will be completely done in the very near future, possibly by the end of this month (April). Expect to see Korean and other languages showing up in the language selection menus on the site as Phonemica expands to be a platform for the documentation of not just Chinese dialects and stories, but of those from around the world. Because after all, everyone has a story to tell.

If you’d like to help us get the site ready to accept recordings in your language, please get in touch with me directly at keyilan@phonemica.net and we’ll get to work.

0 comments

Recording updates: Poem "Saying Good-bye to Cambridge Again" in Lǒngxi Dialect

Recording #297
Volunteer to translate the Chinese version of this blogpost by sending Steve (Steve Hansen 司圆直) an email!

0 comments

recent activity

media coverage

supporters

We thank our anonymous supporters who helped us get the progect off the ground in 2013. We couldn't have done it without you.

If you'd like to help support the project, please drop us a line at info@phonemica.net.

sponsors

3000 Hànzì
Peter Nelson
Jeremy Goldkorn

partners

AllSet Learning
Sinoglot