{ "hq": [ { "speaker": 0, "text": "I've been a principal player to use your the house. That's all.", "start": 0.64, "end": 5.92 }, { "speaker": 1, "text": "Not the house. It's just house.", "start": 5.92, "end": 7.46 }, { "speaker": 0, "text": "Okay. House. Hope so. Thank you. Hey.", "start": 8.41, "end": 20.9525 }, { "speaker": 2, "text": "It's the", "start": 22.050001, "end": 24.147501 }, { "speaker": 1, "text": "second one. Hello,", "start": 25.245, "end": 30.380001 }, { "speaker": 2, "text": "congrats. So you're recording. I'll just talk to you", "start": 32.2, "end": 36.760002 }, { "speaker": 3, "text": "in my office. I mean, it's okay. So", "start": 36.760002, "end": 38.12 }, { "speaker": 2, "text": "That's all I'm just It's", "start": 38.92, "end": 39.8 }, { "speaker": 3, "text": "still going.", "start": 39.8, "end": 40.36 }, { "speaker": 2, "text": "Being in your office for you. So I", "start": 40.440002, "end": 41.72 }, { "speaker": 3, "text": "don't feel this entire life. So, basically, I have a couple of auto squirrels on", "start": 41.72, "end": 45.305 }, { "speaker": 4, "text": "your page.", "start": 45.305, "end": 45.625 }, { "speaker": 1, "text": "So I can differentiate between speakers.", "start": 45.625, "end": 47.325 }, { "speaker": 3, "text": "You all are good at it.", "start": 47.705, "end": 48.585 }, { "speaker": 1, "text": "Fireflies is the best. I'm gonna grab my oh,", "start": 48.585, "end": 50.425 }, { "speaker": 2, "text": "it's really good, definitely.", "start": 50.425, "end": 51.485 }, { "speaker": 4, "text": "That would be choosene at all. I can still fire you if you're not doing your job. I hear from 9", "start": 51.705, "end": 61.039997 }, { "speaker": 2, "text": "to 5.", "start": 61.039997, "end": 61.44 }, { "speaker": 4, "text": "Your billings are practically non existent. What fear? Your request for console. Years behind on your obligations at this point.", "start": 61.44, "end": 70.015 }, { "speaker": 2, "text": "I was right. This doesn't interest me.", "start": 70.015, "end": 72.275 }, { "speaker": 4, "text": "6 years times 3 weeks, better than 4 months.", "start": 72.495, "end": 75.535 }, { "speaker": 2, "text": "It's 5 o'clock. I'm going home. Nice. Look. It's telling.", "start": 75.535, "end": 85.26 }, { "speaker": 4, "text": "Doctor House, the only reason why I don't fire you is because your reputation is still worth something to this house.", "start": 85.26, "end": 90.8 }, { "speaker": 2, "text": "We have a point of agreement not gonna find me.", "start": 90.86, "end": 93.435 }, { "speaker": 4, "text": "Your reputation won't last if you don't do your job. The clinic is part of your job. I want you to do your job.", "start": 93.435, "end": 99.755 }, { "speaker": 2, "text": "This is the last figure one said", "start": 99.755, "end": 101.435 }, { "speaker": 1, "text": "you can't always get somebody used for transcription.", "start": 101.435, "end": 103.855 }, { "speaker": 3, "text": "Perhaps or Deepgram?", "start": 104.98, "end": 106.18 }, { "speaker": 2, "text": "It's", "start": 106.18, "end": 106.340004 }, { "speaker": 1, "text": "just Deepgram. Deepgram Deepgram has diarization, which I believe is actually part of the model. They were able to also diarize whispers, so it's some sort of wrapper speech recognition model. It works it works okay if people are speaking internally. But what I've observed is because it's Splops. Like, they're they're trying to, like, take some media's context. If it misses once it starts missing more until it, like, sort of, like, Fucking back off. It was a", "start": 106.340004, "end": 132.94 }, { "speaker": 3, "text": "Both you are. This hour's speaking. I could Yeah. Yeah. Pick up.", "start": 133.26, "end": 135.9 }, { "speaker": 1, "text": "Like, what's what's it what's it doing now?", "start": 135.9, "end": 137.84 }, { "speaker": 2, "text": "In the interest of saving time and avoiding a lot of boring chitchat later, I'm doctor Greg Rehaas. You can call me Greg. I'm one of 3 doctors staffing this clinic this morning.", "start": 137.98, "end": 146.295 }, { "speaker": 4, "text": "Short, sweet, crap.", "start": 146.455, "end": 147.675 }, { "speaker": 2, "text": "This ray of sunshine is Doctor Lisa Cuddy. Doctor Cuddy runs this whole hospital, so I'm fortunate. It's much too busy to deal with you. I am a board. Okay. As long as schooling bar rolls, I I suppose it's", "start": 147.735, "end": 152.155 }, { "speaker": 3, "text": "you're going to add a schooling bar with over one page so people knows how long it goes.", "start": 158.90001, "end": 163.48001 }, { "speaker": 1, "text": "The scrolling bar is to we're absent", "start": 164.26001, "end": 166.58 }, { "speaker": 2, "text": "because for most of you, this job could be done. I think", "start": 166.58, "end": 168.58 }, { "speaker": 3, "text": "it was that yesterday by school right now.", "start": 168.58, "end": 171.33 }, { "speaker": 2, "text": "If you're particularly annoying, you may speak for this. This is Vicodin. It's mine. Yeah.", "start": 171.375, "end": 176.895 }, { "speaker": 3, "text": "How long kind of schools? No. I", "start": 176.895, "end": 178.335 }, { "speaker": 2, "text": "did not. See, forever? Problem. I had a pain problem. There's one knows. Maybe I'm wrong.", "start": 178.335, "end": 184.125 }, { "speaker": 1, "text": "So If the light's blue is still going, that one's that one had never failed me.", "start": 186.15, "end": 190.40999 }, { "speaker": 0, "text": "Okay.", "start": 190.79, "end": 191.29 }, { "speaker": 2, "text": "And who would rather wait", "start": 191.90999, "end": 193.18999 }, { "speaker": 1, "text": "for this? See the image is that it's here. We can we can stop the record so I can show Andrew.", "start": 193.18999, "end": 197.04 }, { "speaker": 0, "text": "Okay.", "start": 197.04, "end": 197.54 }, { "speaker": 1, "text": "I I generated British Pam.", "start": 200.22499, "end": 201.905 }, { "speaker": 4, "text": "Judy Matthews.", "start": 201.905, "end": 202.23833 }, { "speaker": 2, "text": "So the transcription is always running. Or is it on the phone? Is this 2 hours effort? Yeah. So the", "start": 202.23833, "end": 202.51833 }, { "speaker": 1, "text": "the way the way this works is the transcriptions get these these get the audio gets streamed and back end, then this this is a this is a diarized previous one, but describing the whole file works better because the model has more context So afterwards, the summaries are divided by even transcribing the whole style of the summaries.", "start": 202.51833, "end": 228.183 }, { "speaker": 2, "text": "It is the philosophy to it. Campbell is. So let me see who shows", "start": 228.483, "end": 230.353 } ] }