Is there any way to censor out profanity and foul language in the voice-to-text transcription window in the Unreal Engine chat widget? I have the filter on in Convai playground and it works for interactions, but I’d like to be able to censor anything on screen being said to the AI as well.
Yes, you can handle this on the Unreal Engine side.
A common approach is to iterate through the words in the incoming STT text and check them against a list of blocked terms. If a word matches, you can either censor it or remove it before displaying it in the chat widget.