When using Azure Speech to Text (either with or without a customized model) there are times when the system does not recognize a word. I don't mean the word is wrongly recognized, but that the system acts as if nothing was heard.
Might this be due to the fact that the minimum confidence score hasn't been achieved? I haven't been able to find what the minimum confidence score of Azure Speech to Text is.
Could anyone help me out? Thanks