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
I encountered an issue (large-v3) when transcribing Polish audio. Occasionally, Cyrillic letters appear in the transcription, even though the text should be fully in Polish. For example, instead of "Dzień dobry," the output includes "Dzieнь dobry," where the "ń" is replaced by the Cyrillic character "нь."
I noticed that this problem does not occur when using newly implemented BatchedInferencePipeline using same options.
Tested on master and 1.0.3.
The text was updated successfully, but these errors were encountered:
Hi,
I encountered an issue (large-v3) when transcribing Polish audio. Occasionally, Cyrillic letters appear in the transcription, even though the text should be fully in Polish. For example, instead of "Dzień dobry," the output includes "Dzieнь dobry," where the "ń" is replaced by the Cyrillic character "нь."
I noticed that this problem does not occur when using newly implemented BatchedInferencePipeline using same options.
Tested on master and 1.0.3.
The text was updated successfully, but these errors were encountered: