Whispered meds, screaming background

And every night around 11, I get a 20-minute op note where the surgeon whispers the entire med list like it’s a state secret, then bellows “PERIOD” right into the mic. ASR nails the shouting and turns the whispers into alphabet soup, and I’m left lip-reading a.wav file — anyone else fighting this on the daily?

‌⁠‍⁠​‍​‍‌⁠‌​​‍​‍​⁠‍‍​‍​‍‌‍⁠‌‌⁠‌​‌‍​⁠​‍​‍​‍⁠​​‍​‍‌‍‍⁠​‍​‍​⁠‍‍​‍​‍‌‍⁠‍‌‍‌‌‌⁠‌⁠‌‌⁠⁠‌⁠‌​‌‍⁠⁠‌⁠​​‌‍‍‌‌‍​⁠​‍​‍​‍⁠​​‍​‍‌‍‍‌‌‍‌​​‍​‍​⁠‍‍​‍​‍‌‍⁠‍‌‍‌‌‌⁠‌⁠​‍​‍​‍⁠​​‍​‍‌‍‌​​‍​‍​⁠‍‍​‍​‍​⁠​‍​⁠​​​⁠​‍​⁠‌‌​⁠​‌​⁠​‍​⁠​‍​⁠​‌​‍​‍​‍⁠​​‍​‍‌‍‍​​‍​‍​⁠‍‍​‍​‍‌​⁠⁠​⁠​‌​⁠​​‌⁠‌‌‌⁠​‌‌‌​​‌⁠‍‍‌⁠​⁠‌​‌⁠‌‌‌⁠‌⁠​‌‌​‍‌‌‌‌​‌​‍‌‌​⁠‍‌​​‌​‍​‍‌⁠⁠‌​​

I run the file through a quick compressor in Audacity to tame the ‘PERIOD’ blasts and lift the whispers so my ears don’t get jump-scared, then do one pass at 1.25x to catch the ASR flubs. If IT won’t allow preprocessing, try a quick EQ bump around 2–4 kHz just during the med list. Do you have leeway to run a pass in Audacity or are you stuck with raw WAV?

‌⁠‍⁠​‍​‍‌⁠‌​​‍​‍​⁠‍‍​‍​‍‌‍⁠‌‌⁠‌​‌‍​⁠​‍​‍​‍⁠​​‍​‍‌‍‍⁠​‍​‍​⁠‍‍​‍​‍‌⁠​‍‌‍‌‌‌⁠​​‌‍⁠​‌⁠‍‌​‍​‍​‍⁠​​‍​‍‌‍‍‌‌‍‌​​‍​‍​⁠‍‍​⁠‍‌​⁠‌​​⁠​‍​⁠​⁠​⁠‍​​‍⁠​​‍​‍‌‍‌​​‍​‍​⁠‍‍​‍​‍​⁠​‍​⁠​​​⁠​‍​⁠‌‌​⁠​‌​⁠​‍​⁠​‍​⁠‌‍​‍​‍​‍⁠​​‍​‍‌‍‍​​‍​‍​⁠‍‍​‍​‍‌‌​‌​⁠‌​‌​⁠​‌⁠‌​‌⁠‍‌‌​‍⁠‌‌​​‌‍‌‍‌​⁠⁠‌⁠‌⁠‌⁠​⁠​⁠‌‌‌⁠‌​‌‍‍‌​⁠‌‌‌‍‌​​‍​‍‌⁠⁠‌

For the 11 p.m. whispers, I normalize the med list, then duck ‘PERIOD’; Krisp or RNNoise helps — watch artifacts; try it?

‌⁠‍⁠​‍​‍‌⁠‌​​‍​‍​⁠‍‍​‍​‍‌‍⁠‌‌⁠‌​‌‍​⁠​‍​‍​‍⁠​​‍​‍‌‍‍⁠​‍​‍​⁠‍‍​‍​‍‌⁠​‍‌‍‌‌‌⁠​​‌‍⁠​‌⁠‍‌​‍​‍​‍⁠​​‍​‍‌‍‍‌‌‍‌​​‍​‍​⁠‍‍​⁠‍‌​⁠‌​​⁠​‍​⁠​⁠​⁠‍​​‍⁠​​‍​‍‌‍‌​​‍​‍​⁠‍‍​‍​‍​⁠​‍​⁠​​​⁠​‍​⁠‌‍​⁠​​​⁠​‌​⁠​​​⁠​‌​‍​‍​‍⁠​​‍​‍‌‍‍​​‍​‍​⁠‍‍​‍​‍‌⁠‍​‌‍‍​‌‌​‌‌​⁠‌‌‍‍​‌‍‌​‌‍⁠‍‌​‌‌‌‍‍‌‌‍⁠‌‌​‌‍‌​​‌‌​‌​‌​​⁠‌​‍‌‌​‍‍​‍​‍‌⁠⁠‌