This website uses Cookies. Click Accept to agree to our website's cookie use as described in our Privacy Policy. Click Preferences to customize your cookie settings.
Accept
Reject
Preferences
Log in to ask a question
Log in to ask a question

speech-to-text improvements

Posted on 11-01-2021 09:36 AM
tkirill08
Bronze 1
Bronze 1
Reply posted on --/--/---- --:-- AM

Good afternoon!

I am a user of Speech-to-Text. I use it in order to get a written text from the interviews and courses I shoot myself. After that I correct the text manually.

So, in Russian it works fine, however, 30-40 percents of the words are incorrect. Moreover, there are no Russian punctuation at all.  So I get the speech-to-text transcript, then I create the perfect transcript out of this with correct words and punctuations.

All I want to know is how I can improve Speech-to-Text by using the perfect transcript I have already corrected? Where I can send that data to?

P.S. Sorry for my English. I hope You can understand me 🙂

0 1 466
Topic Labels
0 Likes
1 REPLY 1

Posted on --/--/---- --:-- AM
junlu
Staff
Reply posted on --/--/---- --:-- AM

Hi, 

If you're using the Google Cloud Speech-to-Text API [1] and encounter text quality problem, I would suggest that you can report the issue at the Issue Tracker [2] with the reproduction details for the support to further look into issue with you to improve the quality. 

[1] https://cloud.google.com/speech-to-text
[2] https://cloud.google.com/support/docs/issue-trackers