Monday Friendly Tips to Keep in Mind!

Welcome to the Weekly Friendly Tips post!

Written by fellow CrowdSurfers, these posts are published every Monday and will list the previous weeks most common errors and help with how to avoid making those errors in future tasks!

Monday, September 26, 2016


Hello, CrowdSurfers!


This week’s leader in the FTR Error Board was the issue of words being incorrectly transcribed. Here are just a few examples of what I encountered:

      • Transcribed as: Thank you for having me as a contribution of the team work.
      • Correct Transcription: Thank you for having me as a contributing author to your book.
      • Transcribed as: They could be coming from revet, could be coming from another nervous works model.
      • Correct Transcription: They could be coming from Revit, could be coming from another Navisworks model.


One of the nice things about doing FTRs as opposed to just transcription is that most FTRs are long enough to give us some context about what is being discussed. By focusing on what’s being said throughout the entire piece, we can use what we’ve already heard to help us determine whether what’s been transcribed makes sense with regard to what’s being discussed.


The second most frequent error I encountered this week was random capital letters sprinkled throughout the transcriptions. Remember, an FTR is a stitching together of several individual transcription tasks. As per CrowdSurf guidelines, each of those transcriptions begins with a capital letter, even if it is not the start of a sentence. For this reason, one of our primary jobs in FTR is to spot those capitalized words that end up being mid-sentence in the final, stitched together product and correct them to lowercase.


Hot on the heels of the capitalization issue was timing errors. Remember, if the audio does not match the beginning of the transcription in the yellow (current) box, then the timing is off and needs to be flagged.

Please see this excellent YouTube video for a more in-depth explanation:


And finally, speaker changes. A finished FTR should have speaker change indicators wherever, and only wherever, there is a new human speaker. When reviewing an FTR, it is important to watch for and remove incorrect speaker changes, and also to make sure that there are indicators whenever there is a new speaker.

A very nice explanation of proper speaker changes is included in this YouTube video:


Focusing on these issues should help us all to produce the kind of high-quality work that will keep our CrowdSurf clients coming back to re-fill the queues.


Peace, Love, Transcription!

For more help with Full Text Review tasks, check out the links below: