Datasets:
distil-whisper
/
whisper_transcriptions_greedy_timestamped

Dataset card Data Studio Files Community
file_id
stringlengths
10
247
whisper_transcript
stringlengths
10
3.38k
AMI_EN2001a_H04_MEO069_0233238_0233267
<|0.00|> Mm-hmm.<|0.84|>
AMI_EN2001a_H00_MEE068_0218806_0218816
<|0.00|> with<|5.00|>
AMI_EN2001a_H04_MEO069_0192925_0193128
<|0.00|> then it's probably quicker to just.<|2.28|>
AMI_EN2001a_H01_FEO066_0119776_0119995
<|0.00|> Not the summaries.<|1.08|>
AMI_EN2001a_H02_FEO065_0273775_0273799
<|0.00|> It's.<|0.50|>
AMI_EN2001a_H04_MEO069_0355263_0356005
<|0.00|> I think we might have a lot in common what we calculated, because I,<|2.56|><|2.56|> for my latent semantic analysis, need like counts of words within a document.<|7.08|>
AMI_EN2001a_H04_MEO069_0066569_0067627
<|0.00|> And that would basically be it already, because even the selection...<|2.84|><|2.84|> Would the summaries automatically feed from just how prioritized an individual word or how prioritized an individual utterance is?<|10.44|>
AMI_EN2001a_H01_FEO066_0469418_0469458
<|0.00|> So.<|0.50|>
AMI_EN2001a_H04_MEO069_0376175_0376422
<|0.00|> that they're talking about something different in each different topic segment.<|2.68|>
AMI_EN2001a_H04_MEO069_0463964_0464008
<|0.00|> Yeah.<|0.84|>
AMI_EN2001a_H04_MEO069_0288480_0288760
<|0.00|> because you're making counts for word forms, right?<|1.76|>
AMI_EN2001a_H03_MEE067_0175055_0175131
<|0.00|> Alright, okay.<|2.00|>
AMI_EN2001a_H04_MEO069_0152326_0152938
<|0.00|> so how much they display and starting position at where the,<|3.94|><|3.94|> or maybe the mid position of it.<|5.58|>
AMI_EN2001a_H04_MEO069_0219980_0220267
<|0.00|> You'd probably be quite likely if they're talking about a conference or a person<|2.88|>
AMI_EN2001a_H04_MEO069_0336803_0336835
<|0.00|> Mm-hmm.<|0.84|>
AMI_EN2001a_H04_MEO069_0153515_0153569
<|0.00|> that I don't know.<|0.84|>
AMI_EN2001a_H04_MEO069_0111599_0111834
<|0.00|> Just thinking what's the sim...<|2.00|>
AMI_EN2001a_H04_MEO069_0161372_0161949
<|0.00|> for the information density. And we just tie that to the existing<|3.28|><|3.28|> utterances and tie them to the existing speaker changes.<|5.76|>
AMI_EN2001a_H04_MEO069_0496413_0496766
<|0.00|> I mean, otherwise I can give you the code for loading a dictionary.<|3.36|>
AMI_EN2001a_H01_FEO066_0335843_0336878
<|0.00|> Because I, in my outline I talked about using the discourse acts first.<|9.60|>
AMI_EN2001a_H04_MEO069_0328981_0329934
<|0.00|> they'd probably, they'd like you more if you SSH into another computer,<|4.32|><|4.32|> compress it there and then sort of<|7.00|><|7.00|> copy it into the, into the gateway machine.<|9.44|>
AMI_EN2001a_H04_MEO069_0034882_0035121
<|0.00|> And for altogether, like the display itself.<|2.56|>
AMI_EN2001a_H03_MEE067_0298486_0298558
<|0.00|> I wanted that.<|2.00|>
AMI_EN2001a_H04_MEO069_0006441_0007098
<|0.00|> so that we take the ready-made parts and just see how we get them to work together in the<|4.00|><|4.00|> interface the way we want it, and then we have a working prototype.<|6.60|>
AMI_EN2001a_H04_MEO069_0059865_0059903
<|0.00|> Oh.<|0.84|>
AMI_EN2001a_H00_MEE068_0234692_0234773
<|0.00|> and they're all numbered.<|2.00|>
AMI_EN2001a_H03_MEE067_0381421_0381578
<|0.00|> Well yeah, that is not an issue.<|1.60|>
AMI_EN2001a_H01_FEO066_0467212_0467245
<|0.00|> and you.<|0.50|>
AMI_EN2001a_H03_MEE067_0087250_0088042
<|0.00|> they'll need to be calculated at word level though because otherwise there won't be enough<|4.32|><|4.32|> occurrences of the terms to make any meaningful sense<|7.92|>
AMI_EN2001a_H04_MEO069_0062077_0062113
<|0.00|> OK.<|1.54|>
AMI_EN2001a_H04_MEO069_0035121_0035759
<|0.00|> I think we are easier if it's sitting on the XML<|3.32|><|3.32|> than if it's sitting on the SQL stuff.<|5.04|><|5.04|> Because if it's sitting on the XML.<|6.84|>
AMI_EN2001a_H01_FEO066_0278935_0279426
<|0.00|> And I think it even has its own annotation like digits or something.<|5.00|>
AMI_EN2001a_H04_MEO069_0243376_0243423
<|0.00|> you<|2.06|>
AMI_EN2001a_H03_MEE067_0095821_0095998
<|0.00|> I think we'll have to buffer the audio.<|1.76|>
AMI_EN2001a_H04_MEO069_0512093_0512314
<|0.00|> It's just we are halfway through the project time to have.<|2.20|>
AMI_EN2001a_H04_MEO069_0399135_0399165
<|0.00|> Mm-hmm.<|0.84|>
AMI_EN2001a_H04_MEO069_0060622_0060687
<|0.00|> Mm-hmm.<|0.84|>
AMI_EN2001a_H03_MEE067_0325496_0325559
<|0.00|> Yeah.<|1.54|>
AMI_EN2001a_H01_FEO066_0462660_0463171
<|0.00|> So only words per meeting series.<|5.00|>
AMI_EN2001a_H04_MEO069_0440652_0441151
<|0.00|> that like the highest level of merging it's not the whole XC corpus but individual series.<|4.96|>
AMI_EN2001a_H03_MEE067_0102614_0103261
<|0.00|> it'll just string them all together with like maybe, I don't know, tenths of a second silence<|4.80|><|4.80|> in between each one or something like that.<|6.48|>
AMI_EN2001a_H04_MEO069_0440235_0440252
<|0.00|> I think.<|0.84|>
AMI_EN2001a_H04_MEO069_0069395_0069670
<|0.00|> or are we doing it on world level like the information density calculation?<|3.00|>
AMI_EN2001a_H02_FEO065_0271224_0271427
<|0.00|> yeah, that's just an ID or something.<|2.00|>
AMI_EN2001a_H00_MEE068_0459457_0459476
<|0.00|> Hmm.<|0.50|>
AMI_EN2001a_H00_MEE068_0166346_0166540
<|0.00|> Yeah, I'm pretty sure it's already there.<|2.00|>
AMI_EN2001a_H03_MEE067_0401147_0401387
<|0.00|> file with whatever hierarchical structure you want.<|2.40|>
AMI_EN2001a_H04_MEO069_0257834_0257902
<|0.00|> I'm not biased.<|0.84|>
AMI_EN2001a_H04_MEO069_0110014_0110547
<|0.00|> The other end, I mean, it shouldn't be like,<|1.28|><|1.28|> it should be like 50 megabyte in RAM or something.<|3.28|><|3.28|> It shouldn't be massive, should it?<|5.04|>
AMI_EN2001a_H04_MEO069_0437960_0438134
<|0.00|> Just let's be careful about that because.<|2.00|>
AMI_EN2001a_H03_MEE067_0142298_0142500
<|0.00|> speaker<|2.00|>
AMI_EN2001a_H03_MEE067_0506718_0506755
<|0.00|> Yeah.<|0.84|>
AMI_EN2001a_H04_MEO069_0416339_0416812
<|0.00|> Alternatively, we can probably just make another directory on the beefy scratch space.<|4.72|>
AMI_EN2001a_H04_MEO069_0365657_0365856
<|0.00|> then it's like a more coherent framework.<|2.00|>
AMI_EN2001a_H03_MEE067_0256719_0256803
<|0.00|> So yeah.<|0.84|>
AMI_EN2001a_H01_FEO066_0482572_0482705
<|0.00|> Mm-hmm.<|0.84|>
AMI_EN2001a_H03_MEE067_0156392_0156585
<|0.00|> that everyone's.<|1.00|>
AMI_EN2001a_H01_FEO066_0176870_0177175
<|0.00|> There are time stamps for.<|2.80|>
AMI_EN2001a_H00_MEE068_0163886_0164296
<|0.00|> Yeah, I think those segments for each authoring are split up.<|4.08|>
AMI_EN2001a_H04_MEO069_0423604_0423863
<|0.00|> But if they give us access to it here, sitting on a DICE machine.<|2.60|>
AMI_EN2001a_H00_MEE068_0413663_0413691
<|0.00|> Mm.<|1.54|>
AMI_EN2001a_H01_FEO066_0473933_0473987
<|0.00|> should be<|2.00|>
AMI_EN2001a_H02_FEO065_0485657_0486309
<|0.00|> How long would it take to make the frequency counts with the Java hash table?<|6.44|>
AMI_EN2001a_H03_MEE067_0251604_0251651
<|0.00|> Yeah.<|0.50|>
AMI_EN2001a_H03_MEE067_0145708_0145775
<|0.00|> Yeah.<|0.50|>
AMI_EN2001a_H04_MEO069_0436172_0436254
<|0.00|> Wait, wait, wait.<|0.84|>
AMI_EN2001a_H01_FEO066_0276696_0276792
<|0.00|> Yeah, maybe.<|0.96|>
AMI_EN2001a_H04_MEO069_0139102_0139615
<|0.00|> Yeah, and if you make that structurally very similar to the, the, like one level down.<|5.16|>
AMI_EN2001a_H02_FEO065_0270998_0271224
<|0.00|> Yeah, the XML files.<|2.20|>
AMI_EN2001a_H04_MEO069_0474714_0474803
<|0.00|> That sounds quite reasonable.<|2.00|>
AMI_EN2001a_H04_MEO069_0359202_0359256
<|0.00|> I don't know about that.<|2.00|>
AMI_EN2001a_H00_MEE068_0187080_0187103
<|0.00|> Hmm.<|0.50|>
AMI_EN2001a_H03_MEE067_0069883_0070238
<|0.00|> you've got no way of getting in and extracting just that word.<|2.80|>
AMI_EN2001a_H01_FEO066_0341896_0342034
<|0.00|> So I wouldn't need it.<|1.28|>
AMI_EN2001a_H04_MEO069_0369855_0370933
<|0.00|> Wait, are we using this for the waiting in the end now,<|8.32|><|8.32|> this measure you're calculating?<|10.76|>
AMI_EN2001a_H03_MEE067_0071024_0071050
<|0.00|> Yeah.<|0.50|>
AMI_EN2001a_H04_MEO069_0168067_0168285
<|0.00|> So I guess that would be solvable if not.<|3.24|>
AMI_EN2001a_H01_FEO066_0470162_0470257
<|0.00|> Anymore?<|0.88|>
AMI_EN2001a_H03_MEE067_0174754_0174777
<|0.00|> you<|0.50|>
AMI_EN2001a_H04_MEO069_0137825_0138410
<|0.00|> That means you have to go through every single at-rends in a series of 70 hours of meetings.<|5.84|>
AMI_EN2001a_H04_MEO069_0100150_0100741
<|0.00|> there's still some merit on within utterances cutting out stuff which clearly<|3.42|><|3.42|> isn't relevant at all and that maybe also for the audio we'd have to do.<|5.94|>
AMI_EN2001a_H03_MEE067_0360239_0360740
<|0.00|> well<|1.96|><|1.96|> you need to rule frequency as well<|3.80|>
AMI_EN2001a_H02_FEO065_0273079_0273629
<|0.00|> I think there are quite a lot of numbers in the beginning where there is no time stamp for the numbers.<|5.48|>
AMI_EN2001a_H03_MEE067_0341914_0341971
<|0.00|> So you could just.<|0.84|>
AMI_EN2001a_H04_MEO069_0104613_0104958
<|0.00|> Yeah, I don't know anything about Audion.<|2.64|><|2.64|> I have never seen the players.<|4.24|>
AMI_EN2001a_H04_MEO069_0250042_0250407
<|0.00|> But it would be nice to have some basic system which just displays some stuff.<|3.56|>
AMI_EN2001a_H02_FEO065_0483991_0484027
<|0.00|> Mm-hmm.<|0.84|>
AMI_EN2001a_H02_FEO065_0487980_0488010
<|0.00|> Okay.<|0.50|>
AMI_EN2001a_H00_MEE068_0266349_0266428
<|0.00|> Mm.<|0.84|>
AMI_EN2001a_H02_FEO065_0276303_0276610
<|0.00|> Yeah, in the beginning as well sometimes, I think.<|3.00|>
AMI_EN2001a_H04_MEO069_0253106_0253637
<|0.00|> Does anyone want to sit with me and play for three hours<|3.92|><|3.92|> with NetXML at some point?<|6.48|>
AMI_EN2001a_H04_MEO069_0479888_0480157
<|0.00|> Yes, I'm going to build a dictionary then from that.<|2.08|>
AMI_EN2001a_H04_MEO069_0228234_0229573
<|0.00|> So it seems the data structure isn't a big problem and that basically we don't have to<|5.24|><|5.24|> have all these massive discussions of how we exactly interact with the data structure<|8.40|><|8.40|> because most of our work isn't done with that data structure in memory in the browser.<|13.44|>
AMI_EN2001a_H04_MEO069_0035759_0036968
<|0.00|> We have the Knight XML framework with all its functionality<|4.48|><|4.48|> for synchronizing through all the different levels<|9.48|><|9.74|> whenever there's a change,<|10.62|><|10.62|> whenever something's moving forward and stuff.<|12.92|>
AMI_EN2001a_H04_MEO069_0275512_0275771
<|0.00|> Like, do they look like they are attrances numbers?<|2.40|>
AMI_EN2001a_H03_MEE067_0392676_0393213
<|0.00|> when you click on a segment it's going to have like words or whatever that are important.<|5.20|>
AMI_EN2001a_H04_MEO069_0185119_0185540
<|0.00|> Yeah, and the topics basically they're just an ID probably with a start time or something.<|4.22|>
AMI_EN2001a_H03_MEE067_0088974_0089228
<|0.00|> Yeah, I reckon you can just mean it over the sentence.<|2.52|>
AMI_EN2001a_H01_FEO066_0163329_0163411
<|0.00|> Yeah.<|0.84|>
AMI_EN2001a_H03_MEE067_0509234_0509327
<|0.00|> Yeah, I know.<|0.84|>