0

I used Google cloud SDK Shell to transcribe an audio in Vietnamese. It returns a transcript but the text appears as below. Is there some other command or setting I have to enable to have the correct transcription text?

"type.googleapis.com/google.cloud.speech.v1.LongRunningRecognizeResponse",
 "results": [
   {
     "alternatives": [
       {
         "confidence": 0.95064884,
         "transcript": "kh? ch?u qu? th? n?u m? nh? v?y th? ? ?? k?u v?n ph?ng ?? ho?c l? c?i v?n ph?ng 
          b?c s? ?? l?m sao s? gi?p ch? ?i gi?p cho ch? nha em th? em ngh? l? Ok em ?? ?i 2 l?n r?i```
MinhChau Pham
  • 1
  • 1
    This is a `code page` issue. Save the output to a file and display in an editor that understands Unicode. https://en.wikipedia.org/wiki/Code_page – John Hanley Sep 16 '20 at 23:38

1 Answers1

0

Cloud Shell currently only supports UTF-8, Vietnamese characters are not currently supported by the code page in Cloud Shell. The workaround here is to save your request response in a file (e.g. gcloud ml speech recognize -> transcription.txt) and open it in a text editor that supports the language.

Messier_31
  • 183
  • 6