A Pop Up Archive Guide for KCRW’s #Radio Race contestants!

This post is a special guide for contestants of KCRW’s Radio Race. For more support, consult our full FAQ, or shoot us a question at edison@popuparchive.com 


What is this “Radio Race,” and what does Pop Up Archive have to do with it?

KCRW’s 24-Hour Radio Race is a whirlwind day of high-stakes radio making for producers of any experience level. Radio makers from all over the world will have 24 HOURS to write, record, and edit a nonfiction radio story. At 10AM (PT) on Saturday, all entrants will receive the THEME for this years Race. They will then have 24 hours to create a story that somehow relates to this theme. By Sunday at 10AM PT, their finished piece must be posted on Soundcloud for judging.

With only 24 hours to complete a finished piece, the last thing anyone wants to do is transcribe from scratch. That’s why we’re providing entrants with our auto-transcription toolkit. Use Pop Up Archive to help craft your Radio Race stories, and receive free transcriptions for up to an hour of audio! Sign up for our free demo plan here.



The first thing to do is to upload your audio to a collection on the “My Collections” page. We create one to get you started, but feel free to make as many as you want! Or if you’re using Adobe Audition, you can upload your audio file through the Audition interface using the Pop Up Archive add-on.

Pro tip: All collections are private by default, but you can also choose to make them public, and (optional) back up to Internet Archive. Learn more about privacy options.

*Remember to do this one at a time, unless you want multiple audio files to play on the same Pop Up Archive item page!* Watch this tutorial for more help uploading. 


2) PROCESSING (aka, take a much needed nap while our tech. does the grunt work.)

After upload, wait while we process it for automatic transcripts and tags. You can check on the processing status from the “my collections” page.

Pro tip: Keep in mind that we process audio close to real time (with some extra time depending on server speed.) That means an hour long file will take around an hour to process for a complete auto-transcript. Wav files, however, will take longer; so, convert your files to mp3s and then upload.

Pro tip: Patience, producer grasshopper! Wait until you’ve received the complete transcript notification in your inbox to edit. If you try to edit while the auto-transcript is still processing, it will overwrite your edits.



Our transcripts might need some editing love before you can make good use of them, but we’ve got you covered with our transcription key commands.

Pro tip: Use our transcript editing tools to swiftly clean up your transcripts. Then you can export the text and use it to shape your audio story. You can export transcripts with and without timestamps



Get an overview of the day’s recordings, or zone in on a key moment by searching your audio.

Pro tip: You can search both within a single audio file, or within your entire collection. We’ll search your transcripts for keywords to get to the exact time-stamped point in the audio that you’re looking for:



If you’ve made it this far in the Radio Race, I bet you’ll want to show off your finished piece. We make it easy with an html embed code:

Pro tip: Share your masterpiece with the world! If your audio is public, grab one of our embed codes to post your audio to your own site, or click any of the “share” buttons on the Item page. You can use either a simple player, or an interactive transcript player, featuring tweetable transcript lines! 



Good luck, and happy transcribing!

Guided by Digital Voices

5 unexpected insights from automatic speech recognition


Pop Up Archive has been hard at work implementing new speech recognition software for our partners at organizations like NPR, StoryCorps, and the Hoover Institution. The result of this work means better auto-transcripts, and better auto-transcripts mean better access into hours upon hours of spoken content locked in digital audio.

Along the way, we’ve learned some surprising things about the state of automatic speech recognition. Here’s our crash course in the workings of speech-to-text software:

1. Speech-to-text software learns language like people do.

All automatic speech recognition software learns from whatever data it’s given. So, like a person, the more “well-read” your software is in a particular area, the more it will understand.

2. The human standard for perfect transcription is being questioned.

The gold standard for transcripts has always been human transcription. But as machine learning gets better, a human transcriber won’t necessarily transcribe more accurately than a computer for unfamiliar dialects. Speech-to-text software is trained on many voices, so it can interpret dialects from all over the world. Check out this 2011 Google Tech Talk on “Superhuman speech recognition.

3. Speaking clearly can make you harder to understand.

Since most speech software is trained on naturalistic pronunciations — that is, how you would say a word in a real conversation — speakers that over-articulate may not be properly understood. For example, to clearly pronounce the “t"s in "butter” would go against the Standard American English pronunciation, which is closer to a “d” sound.

4. Not all vocabularies are created equal.

When you create a language model, it’s not just the number of words in the model that contributes to accuracy – it’s how well their distribution matches those of the content. 

5. We’ve only scratched the surface. 

Speaker recognition. Accurate punctuation. Comprehensive geographical and biographical knowledge….

All of these features are not only possible in automatic speech recognition, but will soon be on their way into your own Pop Up Archive auto-transcripts. As we integrate the new software into Pop Up Archive over the next few months, you’ll see  major improvements to our automatic transcription and editing tools. We’ll keep you posted as our new features become available!