You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
It worked for a while but then I tried to change the default model on the server side and it stopped working in Buzz. I reset the values but it didn't solve it, I'm now stuck with this error message instead of the transcription: Error: 'ascii' codec can't encode character '\xf6' in position 8: ordinal not in range(128)
I'm not sure how to give you more information on this bug. Anyhow, I think it could be useful to allow for the creation of custom API models in the settings in order to be able to set custom names. Or even better, fetch the list of models available at the endpoint from v1/models 😊
The text was updated successfully, but these errors were encountered:
I just found the source of the problem! I had a non-supported character in the password field 🙈🙈🙈.
My password was "lösenord" and now that I changed to "losenord", it works again!
I'm so irritated at myself but at the same time, I think this bug should be fixed and if it can't, a warning message should be added if the password has wrong characters!
@PierreMesure Thanks for figuring this out. There have been previous reports of this that had no known cause. I have added a check for non-base64 characters to the Preferences screen. This should make use of Custom APIs more stable.
Hi,
I find it really hard to find a good app to use Whisper on Windows so thank you for making Buzz, it has great potential!
I tried to use it with my own private Whisper server (I use faster-whisper-server)
It worked for a while but then I tried to change the default model on the server side and it stopped working in Buzz. I reset the values but it didn't solve it, I'm now stuck with this error message instead of the transcription:
Error: 'ascii' codec can't encode character '\xf6' in position 8: ordinal not in range(128)
I'm not sure how to give you more information on this bug. Anyhow, I think it could be useful to allow for the creation of custom API models in the settings in order to be able to set custom names. Or even better, fetch the list of models available at the endpoint from
v1/models
😊The text was updated successfully, but these errors were encountered: