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
Get hands-on experience with 20+ free Google Cloud products and $300 in free credit for new customers.
Log in to ask a question

Need Help: "long" & "short" models support for Traditional Chinese (zh-tw) in Speech-to-Text v2

Posted on 03-03-2024 06:17 PM
JonSU
Bronze 1
Bronze 1
Reply posted on --/--/---- --:-- AM

Hello everyone,

I'm currently facing an issue with the Google Cloud Speech-to-Text API v2 and was wondering if anyone here could offer some insight or a workaround. According to the documentation (https://shorturl.at/gAHO9), the "long" model should support the Traditional Chinese language code (zh-tw). However, when I attempt to use this combination, it seems that the "long" model does not actually support zh-tw.

error_message.png

Following is my configuration:
configuration.png

Has anyone else encountered this problem?

8 4 486
Topic Labels
4 REPLIES 4

Posted on --/--/---- --:-- AM
mamyraimov
Bronze 3
Bronze 3
Reply posted on --/--/---- --:-- AM

yes, me for kk-KZ, but I have no answer for your and my question about "long" model in V2

Posted on --/--/---- --:-- AM
JonSU
Bronze 1
Bronze 1
In response to mamyraimov
Reply posted on --/--/---- --:-- AM

After extensive trial and error, I found that only the "chirp" model is compatible with zh-tw. Perhaps you could give it a try.

0 Likes

Posted on --/--/---- --:-- AM
mamyraimov
Bronze 3
Bronze 3
In response to JonSU
Reply posted on --/--/---- --:-- AM

It doesn't work good with kk-kz

Posted on --/--/---- --:-- AM
wcting
Bronze 1
Bronze 1
In response to JonSU
Reply posted on --/--/---- --:-- AM

The model "chirp" does not exist in the location named "global", which region are you using ?