Repository navigation
Replies: 1 comment
|
Adding to a discussion, I see that there are many tools that can do "speech to a text field", but often for languages wilt less speakers we need some custom model or settings that most tools miss. Do you see yourself using a "speech to text field" feature?
I see the Buzz could bring option to configure exactly the model and engine needed for your language. This feature seems to be missing in many tools that provide similar functionality. |
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
Uh oh!
There was an error while loading. Please reload this page.
Would be handy to have windows auto start the app and to choose the miminize behaviour to go to tray rather than the task bar. This will make the watch folder more accessible and less user intrusive.
All reactions