sglanz Posted March 11, 2022 Share Posted March 11, 2022 Attached is an audio clip and a transcription. Let me explain. My audio clip is approx. 3 minutes using Michael's "ITC Translator." No background noise added. It is just picking up whatever is inside the computer. It was set to default parameters. The program uses AI and is documented on the forum. The bumps and hisses are artifacts deriving from the slow processing on my computer. But there are words too. I have been experimenting with speech-to-text, so I ran the recording through Otter.ai . It picks out words when it "hears" them - timestamps not included. The transcript: "S S S S S So, so circle circle circle circle circle circle set set set set set for that's AI for a sec by ach ach is our paper paper or paper or paper or cable sounds sorry sorry SORRY Sorry" If I wanted to, I could attach relevance to this, but that is a subjective judgment. For example, the word "AI" seems appropriate while using an AI program here! "Cable sounds" could refer to a computer cable making the sounds, since there was no external noise source. I don't know about the other words. I don't know what the speech-to-text algorithm is doing. At least it is not guilty of human pareidolia. Maybe just machine pareidolia :) Or, it could be genuine. I invite others to experiment with this technique, with ITC Translator or any other EVP recording, processed (or not) in whatever way you like. You are also welcome to listen to the clip, to see if you pick up the same or different words than the algorithm did, ignoring the bumps and hisses. 20220310-104021.wav 1 Quote Link to comment Share on other sites More sharing options...
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.