elplaguister/DTS_line_datasets · Datasets at Hugging Face

topic_shift bool 2 classes	utterance stringlengths 1 7.9k	session_id stringlengths 7 14
false	but but Yeah , it 's a	QMSum_78
false	Mm - hmm. OK. But then there 's another thing that also Thilo 's involved with , which is , um OK , and and also Da - Dave Gelbart. So there 's this this problem of and w and so we had this meeting. Th - the also Adam , before the the before you went away. Uh we , um regarding the representation of overlaps , because at present , um , because of the limitations of th the interface we 're using , overlaps are , uh , not being encoded by the transcribers in as complete and , uh , detailed a way as it might be , and as might be desired I think would be desired in the corpus ultimately.	QMSum_78
false	Mm - hmm.	QMSum_78
false	So we don't have start and end points at each point where there 's an overlap. We just have the the overlaps encoded in a simple bin. Well , OK. So @ @ the limits of the over of of the interface are such that we were at this meeting we were entertaining how we might either expand the the interface or find other tools which already do what would be useful. Because what would ultimately be , um , ideal in my my view and I think I mean , I had the sense that it was consensus , is that , um , a thorough - going musical score notation would be the best way to go. Because you can have multiple channels , there 's a single time - line , it 's very clear , flexible , and all those nice things.	QMSum_78
false	Mm - hmm.	QMSum_78
false	OK. So , um , um , I spoke I had a meeting with Dave Gelbart on on and he had , uh , excellent ideas on how the interface could be modified to to do this kind of representation. But , um , he in the meantime you were checking into the existence of already , um , existing interfaces which might already have these properties. So , do you wanna say something about that ?	QMSum_78
false	Yes. Um , I talked with , uh , Munich guys from from Ludwi - Ludwig Maximilians University , who do a lot of transcribing and transliterations.	QMSum_78
false	Mm - hmm.	QMSum_78
false	And they basically said they have they have , uh , a tool they developed themselves and they can't give away , uh , f it 's too error - prone , and had it 's not supported , a a a and	QMSum_78
false	Yeah.	QMSum_78
false	But , um , Susanne Bur - Burger , who is at se CMU , he wa who was formally at in Munich and w and is now at with CMU , she said she has something which she uses to do eight channels , uh , trans transliterations , eight channels simultaneously ,	QMSum_78
false	Excuse me.	QMSum_78
false	but it 's running under Windows.	QMSum_78
false	Under Windows.	QMSum_78
false	So I 'm not sure if if if we can use it.	QMSum_78
false	Mm - hmm.	QMSum_78
false	She said she would give it to us.	QMSum_78
false	Mm - hmm.	QMSum_78
false	It wouldn't be a problem. And I 've got some some kind of manual down in my office.	QMSum_78
false	Well , maybe we should get it and if it 's good enough we 'll arrange Windows machines to be available.	QMSum_78
true	Yeah.	QMSum_78
false	Mm - hmm. We could uh , potentially so.	QMSum_78
false	So.	QMSum_78
false	I also wanted to be sure I mean , I 've I 've seen the this this is called Praat , PRAAT , which I guess means spee speech in Dutch or something.	QMSum_78
false	Yep.	QMSum_78
false	Yeah , but then I 'm not sure that 's the right thing for us.	QMSum_78
false	But In terms of it being Windows versus	QMSum_78
false	Yeah.	QMSum_78
false	No , no. Praat isn't Praat 's multi - platform.	QMSum_78
false	But I 'm just wondering , is ?	QMSum_78
false	No. No , Praat Yeah. Yeah.	QMSum_78
false	Oh ! I see.	QMSum_78
false	Yeah.	QMSum_78
false	Oh , I see. So Praat may not be	QMSum_78
false	That 's not Praat. It 's called " trans transedit " I think.	QMSum_78
false	It 's a different one.	QMSum_78
false	The the , uh the tool from from Susanne.	QMSum_78
false	I see. Oh , I see. OK. OK. Alright.	QMSum_78
false	The other thing , uh , to keep in mind , uh I mean , we 've been very concerned to get all this rolling so that we would actually have data ,	QMSum_78
false	Mmm , yeah.	QMSum_78
false	but , um , I think our outside sponsor is actually gonna kick in	QMSum_78
false	Mm - hmm.	QMSum_78
false	and ultimately that path will be smoothed out. So I don't know if we have a long - term need to do lots and lots of transcribing. I think we had a very quick need to get something out and we 'd like to be able to do some later because just it 's inter it 's interesting. But as far a you know , uh , with with any luck we 'll be able to wind down the larger project.	QMSum_78
false	Oh.	QMSum_78
false	But you s	QMSum_78
false	What our decision was is that we 'll go ahead with what we have with a not very fine time scale on the overlaps.	QMSum_78
false	Yeah.	QMSum_78
false	Right. Yeah.	QMSum_78
false	And and do what we can later to clean that up if we need to.	QMSum_78
false	Mm - hmm.	QMSum_78
false	Right.	QMSum_78
false	And and I was just thinking that , um , if it were possible to bring that in , like , you know , this week , then when they 're encoding the overlaps it would be nice for them to be able to specify when you know , the start points and end points of overlaps.	QMSum_78
false	Uh - huh.	QMSum_78
false	uh Th - they 're making really quick progress.	QMSum_78
false	Yeah. That 's great.	QMSum_78
false	And , um , so my my goal was w m my charge was to get eleven hours by the end of the month. And it 'll be I 'm I 'm I 'm clear that we 'll be able to do that.	QMSum_78
false	That 's great.	QMSum_78
false	And did you , uh , forward Morgan Brian 's thing ?	QMSum_78
false	Yeah.	QMSum_78
false	I sent it to , um who did I send that to ? I sent it to a list and I thought I sent it to the e to the local list.	QMSum_78
false	Meeting Recorder.	QMSum_78
false	Oh , you did ? OK. So you probably did get that.	QMSum_78
false	You saw that ? So Brian did tell me that in fact what you said , that , uh that our that they are making progress and that he 's going that they 're going he 's gonna check the f the output of the first transcription and and	QMSum_78
false	I mean , basically it 's it 's all the difference in the world. I mean , basically he 's he 's on it now.	QMSum_78
false	Yeah.	QMSum_78
false	Oh , that 's this is a new development.	QMSum_78
false	So so so this is so i it 'll happen.	QMSum_78
false	OK. Super. Super. OK. Great.	QMSum_78
false	Yeah. I mean , basically it 's just saying that one of our one of our best people is on it ,	QMSum_78
false	Yeah.	QMSum_78
false	you know , who just doesn't happen to be here anymore. Someone else pays him. So	QMSum_78
false	But about the need for transcription ,	QMSum_78
false	Isn't that great ?	QMSum_78
false	I mean , don't we didn't we previously decide that the IBM transcripts would have to be checked anyway and possibly augmented ?	QMSum_78
false	So. Yeah.	QMSum_78
false	Yes. That 's true.	QMSum_78
false	So , I think having a good tool is worth something no matter what.	QMSum_78
false	Mm - hmm.	QMSum_78
false	Yeah. S OK. That 's that 's a good point.	QMSum_78
false	Yeah , and Dave Gelbart did volunteer ,	QMSum_78
false	Good.	QMSum_78
false	and since he 's not here , I 'll repeat it to at least modify Transcriber , which , if we don't have something else that works , I think that 's a pretty good way of going.	QMSum_78
false	Mmm.	QMSum_78
false	Mm - hmm.	QMSum_78
false	And we discussed on some methods to do it. My approach originally , and I 've already hacked on it a little bit it was too slow because I was trying to display all the waveforms. But he pointed out that you don't really have to. I think that 's a good point.	QMSum_78
false	Mm - hmm.	QMSum_78
false	Mm - hmm.	QMSum_78
false	That if you just display the mix waveform and then have a user interface for editing the different channels , that 's perfectly sufficient.	QMSum_78
false	Hmm.	QMSum_78
false	Yeah , exactly. And just keep those things separate. And and , um , Dan Ellis 's hack already allows them to be able to display different waveforms to clarify overlaps and things ,	QMSum_78
false	No. They can only display one ,	QMSum_78
false	so that 's already	QMSum_78
false	but they can listen to different ones.	QMSum_78
false	Oh , yes , but Well , uh , yes , but what I mean is that , uh , from the transcriber 's perspective , uh , those two functions are separate. And Dan Ellis 's hack handles the , um , choice the ability to choose different waveforms from moment to moment.	QMSum_78
false	But only to listen to , not to look at.	QMSum_78
false	Yeah.	QMSum_78
false	Um	QMSum_78
false	The waveform you 're looking at doesn't change.	QMSum_78
false	Yeah.	QMSum_78
false	That 's true.	QMSum_78

false

but but Yeah , it 's a

QMSum_78

false

Mm - hmm. OK. But then there 's another thing that also Thilo 's involved with , which is , um OK , and and also Da - Dave Gelbart. So there 's this this problem of and w and so we had this meeting. Th - the also Adam , before the the before you went away. Uh we , um regarding the representation of overlaps , because at present , um , because of the limitations of th the interface we 're using , overlaps are , uh , not being encoded by the transcribers in as complete and , uh , detailed a way as it might be , and as might be desired I think would be desired in the corpus ultimately.

QMSum_78

false

Mm - hmm.

QMSum_78

false

So we don't have start and end points at each point where there 's an overlap. We just have the the overlaps encoded in a simple bin. Well , OK. So @ @ the limits of the over of of the interface are such that we were at this meeting we were entertaining how we might either expand the the interface or find other tools which already do what would be useful. Because what would ultimately be , um , ideal in my my view and I think I mean , I had the sense that it was consensus , is that , um , a thorough - going musical score notation would be the best way to go. Because you can have multiple channels , there 's a single time - line , it 's very clear , flexible , and all those nice things.

QMSum_78

false

Mm - hmm.

QMSum_78

false

OK. So , um , um , I spoke I had a meeting with Dave Gelbart on on and he had , uh , excellent ideas on how the interface could be modified to to do this kind of representation. But , um , he in the meantime you were checking into the existence of already , um , existing interfaces which might already have these properties. So , do you wanna say something about that ?

QMSum_78

false

Yes. Um , I talked with , uh , Munich guys from from Ludwi - Ludwig Maximilians University , who do a lot of transcribing and transliterations.

QMSum_78

false

Mm - hmm.

QMSum_78

false

And they basically said they have they have , uh , a tool they developed themselves and they can't give away , uh , f it 's too error - prone , and had it 's not supported , a a a and

QMSum_78

false

Yeah.

QMSum_78

false

But , um , Susanne Bur - Burger , who is at se CMU , he wa who was formally at in Munich and w and is now at with CMU , she said she has something which she uses to do eight channels , uh , trans transliterations , eight channels simultaneously ,

QMSum_78

false

Excuse me.

QMSum_78

false

but it 's running under Windows.

QMSum_78

false

Under Windows.

QMSum_78

false

So I 'm not sure if if if we can use it.

QMSum_78

false

Mm - hmm.

QMSum_78

false

She said she would give it to us.

QMSum_78

false

Mm - hmm.

QMSum_78

false

It wouldn't be a problem. And I 've got some some kind of manual down in my office.

QMSum_78

false

Well , maybe we should get it and if it 's good enough we 'll arrange Windows machines to be available.

QMSum_78

true

Yeah.

QMSum_78

false

Mm - hmm. We could uh , potentially so.

QMSum_78

false

So.

QMSum_78

false

I also wanted to be sure I mean , I 've I 've seen the this this is called Praat , PRAAT , which I guess means spee speech in Dutch or something.

QMSum_78

false

Yep.

QMSum_78

false

Yeah , but then I 'm not sure that 's the right thing for us.

QMSum_78

false

But In terms of it being Windows versus

QMSum_78

false

Yeah.

QMSum_78

false

No , no. Praat isn't Praat 's multi - platform.

QMSum_78

false

But I 'm just wondering , is ?

QMSum_78

false

No. No , Praat Yeah. Yeah.

QMSum_78

false

Oh ! I see.

QMSum_78

false

Yeah.

QMSum_78

false

Oh , I see. So Praat may not be

QMSum_78

false

That 's not Praat. It 's called " trans transedit " I think.

QMSum_78

false

It 's a different one.

QMSum_78

false

The the , uh the tool from from Susanne.

QMSum_78

false

I see. Oh , I see. OK. OK. Alright.

QMSum_78

false

The other thing , uh , to keep in mind , uh I mean , we 've been very concerned to get all this rolling so that we would actually have data ,

QMSum_78

false

Mmm , yeah.

QMSum_78

false

but , um , I think our outside sponsor is actually gonna kick in

QMSum_78

false

Mm - hmm.

QMSum_78

false

and ultimately that path will be smoothed out. So I don't know if we have a long - term need to do lots and lots of transcribing. I think we had a very quick need to get something out and we 'd like to be able to do some later because just it 's inter it 's interesting. But as far a you know , uh , with with any luck we 'll be able to wind down the larger project.

QMSum_78

false

Oh.

QMSum_78

false

But you s

QMSum_78

false

What our decision was is that we 'll go ahead with what we have with a not very fine time scale on the overlaps.

QMSum_78

false

Yeah.

QMSum_78

false

Right. Yeah.

QMSum_78

false

And and do what we can later to clean that up if we need to.

QMSum_78

false

Mm - hmm.

QMSum_78

false

Right.

QMSum_78

false

And and I was just thinking that , um , if it were possible to bring that in , like , you know , this week , then when they 're encoding the overlaps it would be nice for them to be able to specify when you know , the start points and end points of overlaps.

QMSum_78

false

Uh - huh.

QMSum_78

false

uh Th - they 're making really quick progress.

QMSum_78

false

Yeah. That 's great.

QMSum_78

false

And , um , so my my goal was w m my charge was to get eleven hours by the end of the month. And it 'll be I 'm I 'm I 'm clear that we 'll be able to do that.

QMSum_78

false

That 's great.

QMSum_78

false

And did you , uh , forward Morgan Brian 's thing ?

QMSum_78

false

Yeah.

QMSum_78

false

I sent it to , um who did I send that to ? I sent it to a list and I thought I sent it to the e to the local list.

QMSum_78

false

Meeting Recorder.

QMSum_78

false

Oh , you did ? OK. So you probably did get that.

QMSum_78

false

You saw that ? So Brian did tell me that in fact what you said , that , uh that our that they are making progress and that he 's going that they 're going he 's gonna check the f the output of the first transcription and and

QMSum_78

false

I mean , basically it 's it 's all the difference in the world. I mean , basically he 's he 's on it now.

QMSum_78

false

Yeah.

QMSum_78

false

Oh , that 's this is a new development.

QMSum_78

false

So so so this is so i it 'll happen.

QMSum_78

false

OK. Super. Super. OK. Great.

QMSum_78

false

Yeah. I mean , basically it 's just saying that one of our one of our best people is on it ,

QMSum_78

false

Yeah.

QMSum_78

false

you know , who just doesn't happen to be here anymore. Someone else pays him. So

QMSum_78

false

But about the need for transcription ,

QMSum_78

false

Isn't that great ?

QMSum_78

false

I mean , don't we didn't we previously decide that the IBM transcripts would have to be checked anyway and possibly augmented ?

QMSum_78

false

So. Yeah.

QMSum_78

false

Yes. That 's true.

QMSum_78

false

So , I think having a good tool is worth something no matter what.

QMSum_78

false

Mm - hmm.

QMSum_78

false

Yeah. S OK. That 's that 's a good point.

QMSum_78

false

Yeah , and Dave Gelbart did volunteer ,

QMSum_78

false

Good.

QMSum_78

false

and since he 's not here , I 'll repeat it to at least modify Transcriber , which , if we don't have something else that works , I think that 's a pretty good way of going.

QMSum_78

false

Mmm.

QMSum_78

false

Mm - hmm.

QMSum_78

false

And we discussed on some methods to do it. My approach originally , and I 've already hacked on it a little bit it was too slow because I was trying to display all the waveforms. But he pointed out that you don't really have to. I think that 's a good point.

QMSum_78

false

Mm - hmm.

QMSum_78

false

Mm - hmm.

QMSum_78

false

That if you just display the mix waveform and then have a user interface for editing the different channels , that 's perfectly sufficient.

QMSum_78

false

Hmm.

QMSum_78

false

Yeah , exactly. And just keep those things separate. And and , um , Dan Ellis 's hack already allows them to be able to display different waveforms to clarify overlaps and things ,

QMSum_78

false

No. They can only display one ,

QMSum_78

false

so that 's already

QMSum_78

false

but they can listen to different ones.

QMSum_78

false

Oh , yes , but Well , uh , yes , but what I mean is that , uh , from the transcriber 's perspective , uh , those two functions are separate. And Dan Ellis 's hack handles the , um , choice the ability to choose different waveforms from moment to moment.

QMSum_78

false

But only to listen to , not to look at.

QMSum_78

false

Yeah.

QMSum_78

false

Um

QMSum_78

false

The waveform you 're looking at doesn't change.

QMSum_78

false

Yeah.

QMSum_78

false

That 's true.

QMSum_78