0

I am working on building an external site to manage transcriptions and am currently trying to fetch a custom vocabulary I have created for AWS Transcribe using the API. My intention is to fetch the list of words, display them in a textbox for the user to edit, and then send the edited list back to AWS. The 'GetVocabulary' action that exists does allow me to request a custom vocabulary but it simply returns:

Response Syntax

{
   "DownloadUri": "string",
   "FailureReason": "string",
   "LanguageCode": "string",
   "LastModifiedTime": number,
   "VocabularyName": "string",
   "VocabularyState": "string"
}

I have tried grabbing/scraping the text document from the "DownloadUri" in many different ways but have not been successful due to receiving all kinds of access/CORS issues. This was something I was able to get around for normal transcription files since they are stored on my own S3 buckets for which I can set the access permissions. Custom vocabularies, however, are stored somewhere on AWS's own default buckets (I have no idea where) and thus I cannot change the access permissions (or at least don't know where to do so).

My current workaround is to simply load the DownloadUri into an iframe and have the user simply copy and paste the words from the iframe into a textbox and then edit them. This is obviously not ideal/elegant at all though.

Any help/info with this would be greatly appreciated, thanks!!

Joey Martin
  • 11
  • 2
  • The documentation says this about `DownloadUri`: "The S3 location where the vocabulary is stored. Use this URI to get the contents of the vocabulary. The URI is available for a limited time." So, I wonder if the problem is that the URI has expired before you are using it? Also, the URI is very long, so make sure it isn't being truncated. (You can view it in the Custom Vocabulary console, but always re-generate it when it is needed.) – John Rotenstein Mar 12 '19 at 21:47
  • @JohnRotenstein Thank you so much for your reply. Unfortunately, neither of these are the issue. I am requesting a new DownloadUri to use every time the user requests a vocabulary so the link is less than a second old when I am trying to access it. As for the matter of truncation, this is also not the issue as I am able to load and display the vocabulary perfectly fine in an iframe on the page. Thanks for the help, but the search continues haha – Joey Martin Mar 13 '19 at 13:27
  • Please provide more details of the error you are receiving so that we can attempt to reproduce the situation. – John Rotenstein Mar 13 '19 at 20:01

0 Answers0