elplaguister/DTS_pair_datasets · Datasets at Hugging Face

prev1 stringlengths 1 7.9k	prev2 stringlengths 1 7.9k ⌀	next1 stringlengths 1 7.9k	next2 stringlengths 1 7.9k ⌀	is_boundary bool 2 classes	session_id stringlengths 7 14
It 's not very significant.	null	Uh , channel one. Yes.	Channel three.	false	QMSum_120
It 's not very significant.	Uh , channel one. Yes.	Channel three.	OK.	false	QMSum_120
Uh , channel one. Yes.	Channel three.	OK.	Mm - hmm.	false	QMSum_120
Channel three.	OK.	Mm - hmm.	Channel three.	false	QMSum_120
OK.	Mm - hmm.	Channel three.	Ta	false	QMSum_120
Mm - hmm.	Channel three.	Ta	Channel three. Alright.	false	QMSum_120
Channel three.	Ta	Channel three. Alright.	OK , did you solve speech recognition last week ?	false	QMSum_120
Ta	Channel three. Alright.	OK , did you solve speech recognition last week ?	Almost.	false	QMSum_120
Channel three. Alright.	OK , did you solve speech recognition last week ?	Almost.	Alright ! Let 's do image processing.	false	QMSum_120
OK , did you solve speech recognition last week ?	Almost.	Alright ! Let 's do image processing.	Yes , again.	false	QMSum_120
Almost.	Alright ! Let 's do image processing.	Yes , again.	Great.	false	QMSum_120
Alright ! Let 's do image processing.	Yes , again.	Great.	We did it again , Morgan.	false	QMSum_120
Yes , again.	Great.	We did it again , Morgan.	Alright !	false	QMSum_120
Great.	We did it again , Morgan.	Alright !	Doo - doop , doo - doo.	false	QMSum_120
We did it again , Morgan.	Alright !	Doo - doop , doo - doo.	What 's wrong with ?	false	QMSum_120
Alright !	Doo - doop , doo - doo.	What 's wrong with ?	OK. It 's April fifth. Actually , Hynek should be getting back in town shortly if he isn't already.	false	QMSum_120
Doo - doop , doo - doo.	What 's wrong with ?	OK. It 's April fifth. Actually , Hynek should be getting back in town shortly if he isn't already.	Is he gonna come here ?	false	QMSum_120
What 's wrong with ?	OK. It 's April fifth. Actually , Hynek should be getting back in town shortly if he isn't already.	Is he gonna come here ?	Uh. Well , we 'll drag him here. I know where he is.	false	QMSum_120
OK. It 's April fifth. Actually , Hynek should be getting back in town shortly if he isn't already.	Is he gonna come here ?	Uh. Well , we 'll drag him here. I know where he is.	So when you said " in town " , you mean Oregon.	false	QMSum_120
Is he gonna come here ?	Uh. Well , we 'll drag him here. I know where he is.	So when you said " in town " , you mean Oregon.	U u u u uh , I meant , you know , this end of the world , yeah , is really what I meant ,	false	QMSum_120
Uh. Well , we 'll drag him here. I know where he is.	So when you said " in town " , you mean Oregon.	U u u u uh , I meant , you know , this end of the world , yeah , is really what I meant ,	Oh.	false	QMSum_120
So when you said " in town " , you mean Oregon.	U u u u uh , I meant , you know , this end of the world , yeah , is really what I meant ,	Oh.	Doo , doo - doo.	false	QMSum_120
U u u u uh , I meant , you know , this end of the world , yeah , is really what I meant ,	Oh.	Doo , doo - doo.	uh , cuz he 's been in Europe.	false	QMSum_120
Oh.	Doo , doo - doo.	uh , cuz he 's been in Europe.	Doo - doo.	false	QMSum_120
Doo , doo - doo.	uh , cuz he 's been in Europe.	Doo - doo.	So.	false	QMSum_120
uh , cuz he 's been in Europe.	Doo - doo.	So.	I have something just fairly brief to report on.	false	QMSum_120
Doo - doo.	So.	I have something just fairly brief to report on.	Mmm.	false	QMSum_120
So.	I have something just fairly brief to report on.	Mmm.	Um , I did some experim uh , uh , just a few more experiments before I had to , uh , go away for the w well , that week.	false	QMSum_120
I have something just fairly brief to report on.	Mmm.	Um , I did some experim uh , uh , just a few more experiments before I had to , uh , go away for the w well , that week.	Great !	false	QMSum_120
Mmm.	Um , I did some experim uh , uh , just a few more experiments before I had to , uh , go away for the w well , that week.	Great !	Was it last week or whenever ? Um , so what I was started playing with was the th again , this is the HTK back - end. And , um , I was curious because the way that they train up the models , they go through about four sort of rounds of of training. And in the first round they do uh , I think it 's three iterations , and for the last three rounds e e they do seven iterations of re - estimation in each of those three. And so , you know , that 's part of what takes so long to train the the the back - end for this.	false	QMSum_120
Um , I did some experim uh , uh , just a few more experiments before I had to , uh , go away for the w well , that week.	Great !	Was it last week or whenever ? Um , so what I was started playing with was the th again , this is the HTK back - end. And , um , I was curious because the way that they train up the models , they go through about four sort of rounds of of training. And in the first round they do uh , I think it 's three iterations , and for the last three rounds e e they do seven iterations of re - estimation in each of those three. And so , you know , that 's part of what takes so long to train the the the back - end for this.	I 'm sorry , I didn't quite get that. There 's there 's four and there 's seven and I I 'm sorry.	false	QMSum_120
Great !	Was it last week or whenever ? Um , so what I was started playing with was the th again , this is the HTK back - end. And , um , I was curious because the way that they train up the models , they go through about four sort of rounds of of training. And in the first round they do uh , I think it 's three iterations , and for the last three rounds e e they do seven iterations of re - estimation in each of those three. And so , you know , that 's part of what takes so long to train the the the back - end for this.	I 'm sorry , I didn't quite get that. There 's there 's four and there 's seven and I I 'm sorry.	Yeah. Uh , maybe I should write it on the board. So , there 's four rounds of training. Um , I g I g I guess you could say iterations. The first one is three , then seven , seven , and seven. And what these numbers refer to is the number of times that the , uh , HMM re - estimation is run. It 's this program called H E	false	QMSum_120
Was it last week or whenever ? Um , so what I was started playing with was the th again , this is the HTK back - end. And , um , I was curious because the way that they train up the models , they go through about four sort of rounds of of training. And in the first round they do uh , I think it 's three iterations , and for the last three rounds e e they do seven iterations of re - estimation in each of those three. And so , you know , that 's part of what takes so long to train the the the back - end for this.	I 'm sorry , I didn't quite get that. There 's there 's four and there 's seven and I I 'm sorry.	Yeah. Uh , maybe I should write it on the board. So , there 's four rounds of training. Um , I g I g I guess you could say iterations. The first one is three , then seven , seven , and seven. And what these numbers refer to is the number of times that the , uh , HMM re - estimation is run. It 's this program called H E	But in HTK , what 's the difference between , uh , a an inner loop and an outer loop in these iterations ?	false	QMSum_120
I 'm sorry , I didn't quite get that. There 's there 's four and there 's seven and I I 'm sorry.	Yeah. Uh , maybe I should write it on the board. So , there 's four rounds of training. Um , I g I g I guess you could say iterations. The first one is three , then seven , seven , and seven. And what these numbers refer to is the number of times that the , uh , HMM re - estimation is run. It 's this program called H E	But in HTK , what 's the difference between , uh , a an inner loop and an outer loop in these iterations ?	OK. So what happens is , um , at each one of these points , you increase the number of Gaussians in the model.	false	QMSum_120
Yeah. Uh , maybe I should write it on the board. So , there 's four rounds of training. Um , I g I g I guess you could say iterations. The first one is three , then seven , seven , and seven. And what these numbers refer to is the number of times that the , uh , HMM re - estimation is run. It 's this program called H E	But in HTK , what 's the difference between , uh , a an inner loop and an outer loop in these iterations ?	OK. So what happens is , um , at each one of these points , you increase the number of Gaussians in the model.	Yeah. Oh , right ! This was the mix up stuff.	false	QMSum_120
But in HTK , what 's the difference between , uh , a an inner loop and an outer loop in these iterations ?	OK. So what happens is , um , at each one of these points , you increase the number of Gaussians in the model.	Yeah. Oh , right ! This was the mix up stuff.	Yeah. The mix up.	false	QMSum_120
OK. So what happens is , um , at each one of these points , you increase the number of Gaussians in the model.	Yeah. Oh , right ! This was the mix up stuff.	Yeah. The mix up.	That 's right.	false	QMSum_120
Yeah. Oh , right ! This was the mix up stuff.	Yeah. The mix up.	That 's right.	Right.	false	QMSum_120
Yeah. The mix up.	That 's right.	Right.	I remember now.	false	QMSum_120
That 's right.	Right.	I remember now.	And so , in the final one here , you end up with , uh for all of the the digit words , you end up with , uh , three mixtures per state ,	false	QMSum_120
Right.	I remember now.	And so , in the final one here , you end up with , uh for all of the the digit words , you end up with , uh , three mixtures per state ,	Yeah.	false	QMSum_120
I remember now.	And so , in the final one here , you end up with , uh for all of the the digit words , you end up with , uh , three mixtures per state ,	Yeah.	eh , in the final thing. So I had done some experiments where I was I I want to play with the number of mixtures.	false	QMSum_120
And so , in the final one here , you end up with , uh for all of the the digit words , you end up with , uh , three mixtures per state ,	Yeah.	eh , in the final thing. So I had done some experiments where I was I I want to play with the number of mixtures.	Mm - hmm.	false	QMSum_120
Yeah.	eh , in the final thing. So I had done some experiments where I was I I want to play with the number of mixtures.	Mm - hmm.	But , um , uh , I wanted to first test to see if we actually need to do this many iterations early on.	false	QMSum_120
eh , in the final thing. So I had done some experiments where I was I I want to play with the number of mixtures.	Mm - hmm.	But , um , uh , I wanted to first test to see if we actually need to do this many iterations early on.	Uh , one , two ,	false	QMSum_120
Mm - hmm.	But , um , uh , I wanted to first test to see if we actually need to do this many iterations early on.	Uh , one , two ,	Mm - hmm.	false	QMSum_120
But , um , uh , I wanted to first test to see if we actually need to do this many iterations early on.	Uh , one , two ,	Mm - hmm.	And so , um , I I ran a couple of experiments where I reduced that to l to be three , two , two , uh , five , I think , and I got almost the exact same results.	false	QMSum_120
Uh , one , two ,	Mm - hmm.	And so , um , I I ran a couple of experiments where I reduced that to l to be three , two , two , uh , five , I think , and I got almost the exact same results.	Mm - hmm.	false	QMSum_120
Mm - hmm.	And so , um , I I ran a couple of experiments where I reduced that to l to be three , two , two , uh , five , I think , and I got almost the exact same results.	Mm - hmm.	And but it runs much much faster. So , um , I I think m it only took something like , uh , three or four hours to do the full training ,	false	QMSum_120
And so , um , I I ran a couple of experiments where I reduced that to l to be three , two , two , uh , five , I think , and I got almost the exact same results.	Mm - hmm.	And but it runs much much faster. So , um , I I think m it only took something like , uh , three or four hours to do the full training ,	As opposed to ?	false	QMSum_120
Mm - hmm.	And but it runs much much faster. So , um , I I think m it only took something like , uh , three or four hours to do the full training ,	As opposed to ?	Good.	false	QMSum_120
And but it runs much much faster. So , um , I I think m it only took something like , uh , three or four hours to do the full training ,	As opposed to ?	Good.	as opposed to wh what , sixteen hours or something like that ? I mean , it takes you have to do an overnight basically , the way it is set up now.	false	QMSum_120
As opposed to ?	Good.	as opposed to wh what , sixteen hours or something like that ? I mean , it takes you have to do an overnight basically , the way it is set up now.	Yeah. It depends.	false	QMSum_120
Good.	as opposed to wh what , sixteen hours or something like that ? I mean , it takes you have to do an overnight basically , the way it is set up now.	Yeah. It depends.	Mm - hmm.	false	QMSum_120
as opposed to wh what , sixteen hours or something like that ? I mean , it takes you have to do an overnight basically , the way it is set up now.	Yeah. It depends.	Mm - hmm.	Mm - hmm.	false	QMSum_120
Yeah. It depends.	Mm - hmm.	Mm - hmm.	So , uh , even we don't do anything else , doing something like this could allow us to turn experiments around a lot faster.	false	QMSum_120
Mm - hmm.	Mm - hmm.	So , uh , even we don't do anything else , doing something like this could allow us to turn experiments around a lot faster.	And then when you have your final thing , do a full one , so it 's	false	QMSum_120
Mm - hmm.	So , uh , even we don't do anything else , doing something like this could allow us to turn experiments around a lot faster.	And then when you have your final thing , do a full one , so it 's	And when you have your final thing , we go back to this.	false	QMSum_120
So , uh , even we don't do anything else , doing something like this could allow us to turn experiments around a lot faster.	And then when you have your final thing , do a full one , so it 's	And when you have your final thing , we go back to this.	Yeah.	false	QMSum_120
And then when you have your final thing , do a full one , so it 's	And when you have your final thing , we go back to this.	Yeah.	So , um , and it 's a real simple change to make. I mean , it 's like one little text file you edit and change those numbers , and you don't do anything else.	false	QMSum_120
And when you have your final thing , we go back to this.	Yeah.	So , um , and it 's a real simple change to make. I mean , it 's like one little text file you edit and change those numbers , and you don't do anything else.	Oh , this is a	false	QMSum_120
Yeah.	So , um , and it 's a real simple change to make. I mean , it 's like one little text file you edit and change those numbers , and you don't do anything else.	Oh , this is a	Mm - hmm.	false	QMSum_120
So , um , and it 's a real simple change to make. I mean , it 's like one little text file you edit and change those numbers , and you don't do anything else.	Oh , this is a	Mm - hmm.	And then you just run.	false	QMSum_120
Oh , this is a	Mm - hmm.	And then you just run.	OK.	false	QMSum_120
Mm - hmm.	And then you just run.	OK.	So it 's a very simple change to make and it doesn't seem to hurt all that much.	false	QMSum_120
And then you just run.	OK.	So it 's a very simple change to make and it doesn't seem to hurt all that much.	So you you run with three , two , two , five ? That 's a	false	QMSum_120
OK.	So it 's a very simple change to make and it doesn't seem to hurt all that much.	So you you run with three , two , two , five ? That 's a	So I Uh , I I have to look to see what the exact numbers were.	false	QMSum_120
So it 's a very simple change to make and it doesn't seem to hurt all that much.	So you you run with three , two , two , five ? That 's a	So I Uh , I I have to look to see what the exact numbers were.	Yeah.	false	QMSum_120
So you you run with three , two , two , five ? That 's a	So I Uh , I I have to look to see what the exact numbers were.	Yeah.	I I thought was , like , three , two , two , five ,	false	QMSum_120
So I Uh , I I have to look to see what the exact numbers were.	Yeah.	I I thought was , like , three , two , two , five ,	Mm - hmm.	false	QMSum_120
Yeah.	I I thought was , like , three , two , two , five ,	Mm - hmm.	but I I 'll I 'll double check. It was over a week ago that I did it ,	false	QMSum_120
I I thought was , like , three , two , two , five ,	Mm - hmm.	but I I 'll I 'll double check. It was over a week ago that I did it ,	OK. Mm - hmm.	false	QMSum_120
Mm - hmm.	but I I 'll I 'll double check. It was over a week ago that I did it ,	OK. Mm - hmm.	so I can't remember exactly.	false	QMSum_120
but I I 'll I 'll double check. It was over a week ago that I did it ,	OK. Mm - hmm.	so I can't remember exactly.	Oh.	false	QMSum_120
OK. Mm - hmm.	so I can't remember exactly.	Oh.	But , uh	false	QMSum_120
so I can't remember exactly.	Oh.	But , uh	Mm - hmm.	false	QMSum_120
Oh.	But , uh	Mm - hmm.	um , but it 's so much faster. I it makes a big difference.	false	QMSum_120
But , uh	Mm - hmm.	um , but it 's so much faster. I it makes a big difference.	Hmm.	false	QMSum_120
Mm - hmm.	um , but it 's so much faster. I it makes a big difference.	Hmm.	So we could do a lot more experiments and throw a lot more stuff in there.	false	QMSum_120
um , but it 's so much faster. I it makes a big difference.	Hmm.	So we could do a lot more experiments and throw a lot more stuff in there.	Yeah.	false	QMSum_120
Hmm.	So we could do a lot more experiments and throw a lot more stuff in there.	Yeah.	That 's great.	false	QMSum_120
So we could do a lot more experiments and throw a lot more stuff in there.	Yeah.	That 's great.	Um. Oh , the other thing that I did was , um , I compiled the HTK stuff for the Linux boxes. So we have this big thing that we got from IBM , which is a five - processor machine. Really fast , but it 's running Linux. So , you can now run your experiments on that machine and you can run five at a time and it runs , uh , as fast as , you know , uh , five different machines.	false	QMSum_120
Yeah.	That 's great.	Um. Oh , the other thing that I did was , um , I compiled the HTK stuff for the Linux boxes. So we have this big thing that we got from IBM , which is a five - processor machine. Really fast , but it 's running Linux. So , you can now run your experiments on that machine and you can run five at a time and it runs , uh , as fast as , you know , uh , five different machines.	Mm - hmm.	false	QMSum_120
That 's great.	Um. Oh , the other thing that I did was , um , I compiled the HTK stuff for the Linux boxes. So we have this big thing that we got from IBM , which is a five - processor machine. Really fast , but it 's running Linux. So , you can now run your experiments on that machine and you can run five at a time and it runs , uh , as fast as , you know , uh , five different machines.	Mm - hmm.	Mm - hmm.	false	QMSum_120
Um. Oh , the other thing that I did was , um , I compiled the HTK stuff for the Linux boxes. So we have this big thing that we got from IBM , which is a five - processor machine. Really fast , but it 's running Linux. So , you can now run your experiments on that machine and you can run five at a time and it runs , uh , as fast as , you know , uh , five different machines.	Mm - hmm.	Mm - hmm.	So , um , I 've forgotten now what the name of that machine is but I can I can send email around about it.	false	QMSum_120
Mm - hmm.	Mm - hmm.	So , um , I 've forgotten now what the name of that machine is but I can I can send email around about it.	Yeah.	false	QMSum_120
Mm - hmm.	So , um , I 've forgotten now what the name of that machine is but I can I can send email around about it.	Yeah.	And so we 've got it now HTK 's compiled for both the Linux and for , um , the Sparcs. Um , you have to make you have to make sure that in your dot CSHRC , um , it detects whether you 're running on the Linux or a a Sparc and points to the right executables. Uh , and you may not have had that in your dot CSHRC before , if you were always just running the Sparc. So , um ,	false	QMSum_120
So , um , I 've forgotten now what the name of that machine is but I can I can send email around about it.	Yeah.	And so we 've got it now HTK 's compiled for both the Linux and for , um , the Sparcs. Um , you have to make you have to make sure that in your dot CSHRC , um , it detects whether you 're running on the Linux or a a Sparc and points to the right executables. Uh , and you may not have had that in your dot CSHRC before , if you were always just running the Sparc. So , um ,	Mm - hmm.	false	QMSum_120
Yeah.	And so we 've got it now HTK 's compiled for both the Linux and for , um , the Sparcs. Um , you have to make you have to make sure that in your dot CSHRC , um , it detects whether you 're running on the Linux or a a Sparc and points to the right executables. Uh , and you may not have had that in your dot CSHRC before , if you were always just running the Sparc. So , um ,	Mm - hmm.	uh , I can I can tell you exactly what you need to do to get all of that to work. But it 'll it really increases what we can run on.	false	QMSum_120
And so we 've got it now HTK 's compiled for both the Linux and for , um , the Sparcs. Um , you have to make you have to make sure that in your dot CSHRC , um , it detects whether you 're running on the Linux or a a Sparc and points to the right executables. Uh , and you may not have had that in your dot CSHRC before , if you were always just running the Sparc. So , um ,	Mm - hmm.	uh , I can I can tell you exactly what you need to do to get all of that to work. But it 'll it really increases what we can run on.	Hmm. Cool.	false	QMSum_120
Mm - hmm.	uh , I can I can tell you exactly what you need to do to get all of that to work. But it 'll it really increases what we can run on.	Hmm. Cool.	So , together with the fact that we 've got these faster Linux boxes and that it takes less time to do these , um , we should be able to crank through a lot more experiments.	false	QMSum_120
uh , I can I can tell you exactly what you need to do to get all of that to work. But it 'll it really increases what we can run on.	Hmm. Cool.	So , together with the fact that we 've got these faster Linux boxes and that it takes less time to do these , um , we should be able to crank through a lot more experiments.	Mm - hmm.	false	QMSum_120
Hmm. Cool.	So , together with the fact that we 've got these faster Linux boxes and that it takes less time to do these , um , we should be able to crank through a lot more experiments.	Mm - hmm.	So.	false	QMSum_120
So , together with the fact that we 've got these faster Linux boxes and that it takes less time to do these , um , we should be able to crank through a lot more experiments.	Mm - hmm.	So.	Hmm.	false	QMSum_120
Mm - hmm.	So.	Hmm.	So after I did that , then what I wanted to do was try increasing the number of mixtures , just to see , um see how how that affects performance.	false	QMSum_120
So.	Hmm.	So after I did that , then what I wanted to do was try increasing the number of mixtures , just to see , um see how how that affects performance.	Yeah.	false	QMSum_120
Hmm.	So after I did that , then what I wanted to do was try increasing the number of mixtures , just to see , um see how how that affects performance.	Yeah.	So.	false	QMSum_120
So after I did that , then what I wanted to do was try increasing the number of mixtures , just to see , um see how how that affects performance.	Yeah.	So.	Yeah. In fact , you could do something like keep exactly the same procedure and then add a fifth thing onto it	false	QMSum_120
Yeah.	So.	Yeah. In fact , you could do something like keep exactly the same procedure and then add a fifth thing onto it	Mm - hmm.	false	QMSum_120
So.	Yeah. In fact , you could do something like keep exactly the same procedure and then add a fifth thing onto it	Mm - hmm.	that had more.	false	QMSum_120

It 's not very significant.

null

Uh , channel one. Yes.

Channel three.

false

QMSum_120

It 's not very significant.

Uh , channel one. Yes.

Channel three.

OK.

false

QMSum_120

Uh , channel one. Yes.

Channel three.

OK.

Mm - hmm.

false

QMSum_120

Channel three.

OK.

Mm - hmm.

Channel three.

false

QMSum_120

OK.

Mm - hmm.

Channel three.

Ta

false

QMSum_120

Mm - hmm.

Channel three.

Ta

Channel three. Alright.

false

QMSum_120

Channel three.

Ta

Channel three. Alright.

OK , did you solve speech recognition last week ?

false

QMSum_120

Ta

Channel three. Alright.

OK , did you solve speech recognition last week ?

Almost.

false

QMSum_120

Channel three. Alright.

OK , did you solve speech recognition last week ?

Almost.

Alright ! Let 's do image processing.

false

QMSum_120

OK , did you solve speech recognition last week ?

Almost.

Alright ! Let 's do image processing.

Yes , again.

false

QMSum_120

Almost.

Alright ! Let 's do image processing.

Yes , again.

Great.

false

QMSum_120

Alright ! Let 's do image processing.

Yes , again.

Great.

We did it again , Morgan.

false

QMSum_120

Yes , again.

Great.

We did it again , Morgan.

Alright !

false

QMSum_120

Great.

We did it again , Morgan.

Alright !

Doo - doop , doo - doo.

false

QMSum_120

We did it again , Morgan.

Alright !

Doo - doop , doo - doo.

What 's wrong with ?

false

QMSum_120

Alright !

Doo - doop , doo - doo.

What 's wrong with ?

OK. It 's April fifth. Actually , Hynek should be getting back in town shortly if he isn't already.

false

QMSum_120

Doo - doop , doo - doo.

What 's wrong with ?

OK. It 's April fifth. Actually , Hynek should be getting back in town shortly if he isn't already.

Is he gonna come here ?

false

QMSum_120

What 's wrong with ?

OK. It 's April fifth. Actually , Hynek should be getting back in town shortly if he isn't already.

Is he gonna come here ?

Uh. Well , we 'll drag him here. I know where he is.

false

QMSum_120

OK. It 's April fifth. Actually , Hynek should be getting back in town shortly if he isn't already.

Is he gonna come here ?

Uh. Well , we 'll drag him here. I know where he is.

So when you said " in town " , you mean Oregon.

false

QMSum_120

Is he gonna come here ?

Uh. Well , we 'll drag him here. I know where he is.

So when you said " in town " , you mean Oregon.

U u u u uh , I meant , you know , this end of the world , yeah , is really what I meant ,

false

QMSum_120

Uh. Well , we 'll drag him here. I know where he is.

So when you said " in town " , you mean Oregon.

U u u u uh , I meant , you know , this end of the world , yeah , is really what I meant ,

Oh.

false

QMSum_120

So when you said " in town " , you mean Oregon.

U u u u uh , I meant , you know , this end of the world , yeah , is really what I meant ,

Oh.

Doo , doo - doo.

false

QMSum_120

U u u u uh , I meant , you know , this end of the world , yeah , is really what I meant ,

Oh.

Doo , doo - doo.

uh , cuz he 's been in Europe.

false

QMSum_120

Oh.

Doo , doo - doo.

uh , cuz he 's been in Europe.

Doo - doo.

false

QMSum_120

Doo , doo - doo.

uh , cuz he 's been in Europe.

Doo - doo.

So.

false

QMSum_120

uh , cuz he 's been in Europe.

Doo - doo.

So.

I have something just fairly brief to report on.

false

QMSum_120

Doo - doo.

So.

I have something just fairly brief to report on.

Mmm.

false

QMSum_120

So.

I have something just fairly brief to report on.

Mmm.

Um , I did some experim uh , uh , just a few more experiments before I had to , uh , go away for the w well , that week.

false

QMSum_120

I have something just fairly brief to report on.

Mmm.

Um , I did some experim uh , uh , just a few more experiments before I had to , uh , go away for the w well , that week.

Great !

false

QMSum_120

Mmm.

Um , I did some experim uh , uh , just a few more experiments before I had to , uh , go away for the w well , that week.

Great !

Was it last week or whenever ? Um , so what I was started playing with was the th again , this is the HTK back - end. And , um , I was curious because the way that they train up the models , they go through about four sort of rounds of of training. And in the first round they do uh , I think it 's three iterations , and for the last three rounds e e they do seven iterations of re - estimation in each of those three. And so , you know , that 's part of what takes so long to train the the the back - end for this.

false

QMSum_120

Um , I did some experim uh , uh , just a few more experiments before I had to , uh , go away for the w well , that week.

Great !

Was it last week or whenever ? Um , so what I was started playing with was the th again , this is the HTK back - end. And , um , I was curious because the way that they train up the models , they go through about four sort of rounds of of training. And in the first round they do uh , I think it 's three iterations , and for the last three rounds e e they do seven iterations of re - estimation in each of those three. And so , you know , that 's part of what takes so long to train the the the back - end for this.

I 'm sorry , I didn't quite get that. There 's there 's four and there 's seven and I I 'm sorry.

false

QMSum_120

Great !

Was it last week or whenever ? Um , so what I was started playing with was the th again , this is the HTK back - end. And , um , I was curious because the way that they train up the models , they go through about four sort of rounds of of training. And in the first round they do uh , I think it 's three iterations , and for the last three rounds e e they do seven iterations of re - estimation in each of those three. And so , you know , that 's part of what takes so long to train the the the back - end for this.

I 'm sorry , I didn't quite get that. There 's there 's four and there 's seven and I I 'm sorry.

Yeah. Uh , maybe I should write it on the board. So , there 's four rounds of training. Um , I g I g I guess you could say iterations. The first one is three , then seven , seven , and seven. And what these numbers refer to is the number of times that the , uh , HMM re - estimation is run. It 's this program called H E

false

QMSum_120

Was it last week or whenever ? Um , so what I was started playing with was the th again , this is the HTK back - end. And , um , I was curious because the way that they train up the models , they go through about four sort of rounds of of training. And in the first round they do uh , I think it 's three iterations , and for the last three rounds e e they do seven iterations of re - estimation in each of those three. And so , you know , that 's part of what takes so long to train the the the back - end for this.

I 'm sorry , I didn't quite get that. There 's there 's four and there 's seven and I I 'm sorry.

Yeah. Uh , maybe I should write it on the board. So , there 's four rounds of training. Um , I g I g I guess you could say iterations. The first one is three , then seven , seven , and seven. And what these numbers refer to is the number of times that the , uh , HMM re - estimation is run. It 's this program called H E

But in HTK , what 's the difference between , uh , a an inner loop and an outer loop in these iterations ?

false

QMSum_120

I 'm sorry , I didn't quite get that. There 's there 's four and there 's seven and I I 'm sorry.

Yeah. Uh , maybe I should write it on the board. So , there 's four rounds of training. Um , I g I g I guess you could say iterations. The first one is three , then seven , seven , and seven. And what these numbers refer to is the number of times that the , uh , HMM re - estimation is run. It 's this program called H E

But in HTK , what 's the difference between , uh , a an inner loop and an outer loop in these iterations ?

OK. So what happens is , um , at each one of these points , you increase the number of Gaussians in the model.

false

QMSum_120

Yeah. Uh , maybe I should write it on the board. So , there 's four rounds of training. Um , I g I g I guess you could say iterations. The first one is three , then seven , seven , and seven. And what these numbers refer to is the number of times that the , uh , HMM re - estimation is run. It 's this program called H E

But in HTK , what 's the difference between , uh , a an inner loop and an outer loop in these iterations ?

OK. So what happens is , um , at each one of these points , you increase the number of Gaussians in the model.

Yeah. Oh , right ! This was the mix up stuff.

false

QMSum_120

But in HTK , what 's the difference between , uh , a an inner loop and an outer loop in these iterations ?

OK. So what happens is , um , at each one of these points , you increase the number of Gaussians in the model.

Yeah. Oh , right ! This was the mix up stuff.

Yeah. The mix up.

false

QMSum_120

OK. So what happens is , um , at each one of these points , you increase the number of Gaussians in the model.

Yeah. Oh , right ! This was the mix up stuff.

Yeah. The mix up.

That 's right.

false

QMSum_120

Yeah. Oh , right ! This was the mix up stuff.

Yeah. The mix up.

That 's right.

Right.

false

QMSum_120

Yeah. The mix up.

That 's right.

Right.

I remember now.

false

QMSum_120

That 's right.

Right.

I remember now.

And so , in the final one here , you end up with , uh for all of the the digit words , you end up with , uh , three mixtures per state ,

false

QMSum_120

Right.

I remember now.

And so , in the final one here , you end up with , uh for all of the the digit words , you end up with , uh , three mixtures per state ,

Yeah.

false

QMSum_120

I remember now.

And so , in the final one here , you end up with , uh for all of the the digit words , you end up with , uh , three mixtures per state ,

Yeah.

eh , in the final thing. So I had done some experiments where I was I I want to play with the number of mixtures.

false

QMSum_120

And so , in the final one here , you end up with , uh for all of the the digit words , you end up with , uh , three mixtures per state ,

Yeah.

eh , in the final thing. So I had done some experiments where I was I I want to play with the number of mixtures.

Mm - hmm.

false

QMSum_120

Yeah.

eh , in the final thing. So I had done some experiments where I was I I want to play with the number of mixtures.

Mm - hmm.

But , um , uh , I wanted to first test to see if we actually need to do this many iterations early on.

false

QMSum_120

eh , in the final thing. So I had done some experiments where I was I I want to play with the number of mixtures.

Mm - hmm.

But , um , uh , I wanted to first test to see if we actually need to do this many iterations early on.

Uh , one , two ,

false

QMSum_120

Mm - hmm.

But , um , uh , I wanted to first test to see if we actually need to do this many iterations early on.

Uh , one , two ,

Mm - hmm.

false

QMSum_120

But , um , uh , I wanted to first test to see if we actually need to do this many iterations early on.

Uh , one , two ,

Mm - hmm.

And so , um , I I ran a couple of experiments where I reduced that to l to be three , two , two , uh , five , I think , and I got almost the exact same results.

false

QMSum_120

Uh , one , two ,

Mm - hmm.

And so , um , I I ran a couple of experiments where I reduced that to l to be three , two , two , uh , five , I think , and I got almost the exact same results.

Mm - hmm.

false

QMSum_120

Mm - hmm.

And so , um , I I ran a couple of experiments where I reduced that to l to be three , two , two , uh , five , I think , and I got almost the exact same results.

Mm - hmm.

And but it runs much much faster. So , um , I I think m it only took something like , uh , three or four hours to do the full training ,

false

QMSum_120

And so , um , I I ran a couple of experiments where I reduced that to l to be three , two , two , uh , five , I think , and I got almost the exact same results.

Mm - hmm.

And but it runs much much faster. So , um , I I think m it only took something like , uh , three or four hours to do the full training ,

As opposed to ?

false

QMSum_120

Mm - hmm.

And but it runs much much faster. So , um , I I think m it only took something like , uh , three or four hours to do the full training ,

As opposed to ?

Good.

false

QMSum_120

And but it runs much much faster. So , um , I I think m it only took something like , uh , three or four hours to do the full training ,

As opposed to ?

Good.

as opposed to wh what , sixteen hours or something like that ? I mean , it takes you have to do an overnight basically , the way it is set up now.

false

QMSum_120

As opposed to ?

Good.

as opposed to wh what , sixteen hours or something like that ? I mean , it takes you have to do an overnight basically , the way it is set up now.

Yeah. It depends.

false

QMSum_120

Good.

as opposed to wh what , sixteen hours or something like that ? I mean , it takes you have to do an overnight basically , the way it is set up now.

Yeah. It depends.

Mm - hmm.

false

QMSum_120

as opposed to wh what , sixteen hours or something like that ? I mean , it takes you have to do an overnight basically , the way it is set up now.

Yeah. It depends.

Mm - hmm.

false

QMSum_120

Yeah. It depends.

Mm - hmm.

So , uh , even we don't do anything else , doing something like this could allow us to turn experiments around a lot faster.

false

QMSum_120

Mm - hmm.

So , uh , even we don't do anything else , doing something like this could allow us to turn experiments around a lot faster.

And then when you have your final thing , do a full one , so it 's

false

QMSum_120

Mm - hmm.

So , uh , even we don't do anything else , doing something like this could allow us to turn experiments around a lot faster.

And then when you have your final thing , do a full one , so it 's

And when you have your final thing , we go back to this.

false

QMSum_120

So , uh , even we don't do anything else , doing something like this could allow us to turn experiments around a lot faster.

And then when you have your final thing , do a full one , so it 's

And when you have your final thing , we go back to this.

Yeah.

false

QMSum_120

And then when you have your final thing , do a full one , so it 's

And when you have your final thing , we go back to this.

Yeah.

So , um , and it 's a real simple change to make. I mean , it 's like one little text file you edit and change those numbers , and you don't do anything else.

false

QMSum_120

And when you have your final thing , we go back to this.

Yeah.

So , um , and it 's a real simple change to make. I mean , it 's like one little text file you edit and change those numbers , and you don't do anything else.

Oh , this is a

false

QMSum_120

Yeah.

So , um , and it 's a real simple change to make. I mean , it 's like one little text file you edit and change those numbers , and you don't do anything else.

Oh , this is a

Mm - hmm.

false

QMSum_120

So , um , and it 's a real simple change to make. I mean , it 's like one little text file you edit and change those numbers , and you don't do anything else.

Oh , this is a

Mm - hmm.

And then you just run.

false

QMSum_120

Oh , this is a

Mm - hmm.

And then you just run.

OK.

false

QMSum_120

Mm - hmm.

And then you just run.

OK.

So it 's a very simple change to make and it doesn't seem to hurt all that much.

false

QMSum_120

And then you just run.

OK.

So it 's a very simple change to make and it doesn't seem to hurt all that much.

So you you run with three , two , two , five ? That 's a

false

QMSum_120

OK.

So it 's a very simple change to make and it doesn't seem to hurt all that much.

So you you run with three , two , two , five ? That 's a

So I Uh , I I have to look to see what the exact numbers were.

false

QMSum_120

So it 's a very simple change to make and it doesn't seem to hurt all that much.

So you you run with three , two , two , five ? That 's a

So I Uh , I I have to look to see what the exact numbers were.

Yeah.

false

QMSum_120

So you you run with three , two , two , five ? That 's a

So I Uh , I I have to look to see what the exact numbers were.

Yeah.

I I thought was , like , three , two , two , five ,

false

QMSum_120

So I Uh , I I have to look to see what the exact numbers were.

Yeah.

I I thought was , like , three , two , two , five ,

Mm - hmm.

false

QMSum_120

Yeah.

I I thought was , like , three , two , two , five ,

Mm - hmm.

but I I 'll I 'll double check. It was over a week ago that I did it ,

false

QMSum_120

I I thought was , like , three , two , two , five ,

Mm - hmm.

but I I 'll I 'll double check. It was over a week ago that I did it ,

OK. Mm - hmm.

false

QMSum_120

Mm - hmm.

but I I 'll I 'll double check. It was over a week ago that I did it ,

OK. Mm - hmm.

so I can't remember exactly.

false

QMSum_120

but I I 'll I 'll double check. It was over a week ago that I did it ,

OK. Mm - hmm.

so I can't remember exactly.

Oh.

false

QMSum_120

OK. Mm - hmm.

so I can't remember exactly.

Oh.

But , uh

false

QMSum_120

so I can't remember exactly.

Oh.

But , uh

Mm - hmm.

false

QMSum_120

Oh.

But , uh

Mm - hmm.

um , but it 's so much faster. I it makes a big difference.

false

QMSum_120

But , uh

Mm - hmm.

um , but it 's so much faster. I it makes a big difference.

Hmm.

false

QMSum_120

Mm - hmm.

um , but it 's so much faster. I it makes a big difference.

Hmm.

So we could do a lot more experiments and throw a lot more stuff in there.

false

QMSum_120

um , but it 's so much faster. I it makes a big difference.

Hmm.

So we could do a lot more experiments and throw a lot more stuff in there.

Yeah.

false

QMSum_120

Hmm.

So we could do a lot more experiments and throw a lot more stuff in there.

Yeah.

That 's great.

false

QMSum_120

So we could do a lot more experiments and throw a lot more stuff in there.

Yeah.

That 's great.

Um. Oh , the other thing that I did was , um , I compiled the HTK stuff for the Linux boxes. So we have this big thing that we got from IBM , which is a five - processor machine. Really fast , but it 's running Linux. So , you can now run your experiments on that machine and you can run five at a time and it runs , uh , as fast as , you know , uh , five different machines.

false

QMSum_120

Yeah.

That 's great.

Um. Oh , the other thing that I did was , um , I compiled the HTK stuff for the Linux boxes. So we have this big thing that we got from IBM , which is a five - processor machine. Really fast , but it 's running Linux. So , you can now run your experiments on that machine and you can run five at a time and it runs , uh , as fast as , you know , uh , five different machines.

Mm - hmm.

false

QMSum_120

That 's great.

Um. Oh , the other thing that I did was , um , I compiled the HTK stuff for the Linux boxes. So we have this big thing that we got from IBM , which is a five - processor machine. Really fast , but it 's running Linux. So , you can now run your experiments on that machine and you can run five at a time and it runs , uh , as fast as , you know , uh , five different machines.

Mm - hmm.

false

QMSum_120

Um. Oh , the other thing that I did was , um , I compiled the HTK stuff for the Linux boxes. So we have this big thing that we got from IBM , which is a five - processor machine. Really fast , but it 's running Linux. So , you can now run your experiments on that machine and you can run five at a time and it runs , uh , as fast as , you know , uh , five different machines.

Mm - hmm.

So , um , I 've forgotten now what the name of that machine is but I can I can send email around about it.

false

QMSum_120

Mm - hmm.

So , um , I 've forgotten now what the name of that machine is but I can I can send email around about it.

Yeah.

false

QMSum_120

Mm - hmm.

So , um , I 've forgotten now what the name of that machine is but I can I can send email around about it.

Yeah.

And so we 've got it now HTK 's compiled for both the Linux and for , um , the Sparcs. Um , you have to make you have to make sure that in your dot CSHRC , um , it detects whether you 're running on the Linux or a a Sparc and points to the right executables. Uh , and you may not have had that in your dot CSHRC before , if you were always just running the Sparc. So , um ,

false

QMSum_120

So , um , I 've forgotten now what the name of that machine is but I can I can send email around about it.

Yeah.

And so we 've got it now HTK 's compiled for both the Linux and for , um , the Sparcs. Um , you have to make you have to make sure that in your dot CSHRC , um , it detects whether you 're running on the Linux or a a Sparc and points to the right executables. Uh , and you may not have had that in your dot CSHRC before , if you were always just running the Sparc. So , um ,

Mm - hmm.

false

QMSum_120

Yeah.

And so we 've got it now HTK 's compiled for both the Linux and for , um , the Sparcs. Um , you have to make you have to make sure that in your dot CSHRC , um , it detects whether you 're running on the Linux or a a Sparc and points to the right executables. Uh , and you may not have had that in your dot CSHRC before , if you were always just running the Sparc. So , um ,

Mm - hmm.

uh , I can I can tell you exactly what you need to do to get all of that to work. But it 'll it really increases what we can run on.

false

QMSum_120

And so we 've got it now HTK 's compiled for both the Linux and for , um , the Sparcs. Um , you have to make you have to make sure that in your dot CSHRC , um , it detects whether you 're running on the Linux or a a Sparc and points to the right executables. Uh , and you may not have had that in your dot CSHRC before , if you were always just running the Sparc. So , um ,

Mm - hmm.

uh , I can I can tell you exactly what you need to do to get all of that to work. But it 'll it really increases what we can run on.

Hmm. Cool.

false

QMSum_120

Mm - hmm.

uh , I can I can tell you exactly what you need to do to get all of that to work. But it 'll it really increases what we can run on.

Hmm. Cool.

So , together with the fact that we 've got these faster Linux boxes and that it takes less time to do these , um , we should be able to crank through a lot more experiments.

false

QMSum_120

uh , I can I can tell you exactly what you need to do to get all of that to work. But it 'll it really increases what we can run on.

Hmm. Cool.

So , together with the fact that we 've got these faster Linux boxes and that it takes less time to do these , um , we should be able to crank through a lot more experiments.

Mm - hmm.

false

QMSum_120

Hmm. Cool.

So , together with the fact that we 've got these faster Linux boxes and that it takes less time to do these , um , we should be able to crank through a lot more experiments.

Mm - hmm.

So.

false

QMSum_120

So , together with the fact that we 've got these faster Linux boxes and that it takes less time to do these , um , we should be able to crank through a lot more experiments.

Mm - hmm.

So.

Hmm.

false

QMSum_120

Mm - hmm.

So.

Hmm.

So after I did that , then what I wanted to do was try increasing the number of mixtures , just to see , um see how how that affects performance.

false

QMSum_120

So.

Hmm.

So after I did that , then what I wanted to do was try increasing the number of mixtures , just to see , um see how how that affects performance.

Yeah.

false

QMSum_120

Hmm.

So after I did that , then what I wanted to do was try increasing the number of mixtures , just to see , um see how how that affects performance.

Yeah.

So.

false

QMSum_120

So after I did that , then what I wanted to do was try increasing the number of mixtures , just to see , um see how how that affects performance.

Yeah.

So.

Yeah. In fact , you could do something like keep exactly the same procedure and then add a fifth thing onto it

false

QMSum_120

Yeah.

So.

Yeah. In fact , you could do something like keep exactly the same procedure and then add a fifth thing onto it

Mm - hmm.

false

QMSum_120

So.

Yeah. In fact , you could do something like keep exactly the same procedure and then add a fifth thing onto it

Mm - hmm.

that had more.

false

QMSum_120