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
Well we 've done that for one meeting.	so. For forced alignment.	Uh oh oh f not for words	Ye - right.	false	QMSum_225
so. For forced alignment.	Uh oh oh f not for words	Ye - right.	I 'm sorry just for overlaps is we did it for not not for words.	false	QMSum_225
Uh oh oh f not for words	Ye - right.	I 'm sorry just for overlaps is we did it for not not for words.	Right. So this would be like if you take the words um you know and force align them on all the individual close talk uh close talking mikes then how good are these sort of in reality	false	QMSum_225
Ye - right.	I 'm sorry just for overlaps is we did it for not not for words.	Right. So this would be like if you take the words um you know and force align them on all the individual close talk uh close talking mikes then how good are these sort of in reality	Right.	false	QMSum_225
I 'm sorry just for overlaps is we did it for not not for words.	Right. So this would be like if you take the words um you know and force align them on all the individual close talk uh close talking mikes then how good are these sort of in reality	Right.	and then I was thinking it	false	QMSum_225
Right. So this would be like if you take the words um you know and force align them on all the individual close talk uh close talking mikes then how good are these sort of in reality	Right.	and then I was thinking it	So we might want to take twenty minutes and do a closer word level transcription. Maybe actually mark the word boundaries.	false	QMSum_225
Right.	and then I was thinking it	So we might want to take twenty minutes and do a closer word level transcription. Maybe actually mark the word boundaries.	Oh or i have someone look at the alignments uh maybe a linguist who can say um you know roughly if these are OK and how far away they are.	false	QMSum_225
and then I was thinking it	So we might want to take twenty minutes and do a closer word level transcription. Maybe actually mark the word boundaries.	Oh or i have someone look at the alignments uh maybe a linguist who can say um you know roughly if these are OK and how far away they are.	Yeah.	false	QMSum_225
So we might want to take twenty minutes and do a closer word level transcription. Maybe actually mark the word boundaries.	Oh or i have someone look at the alignments uh maybe a linguist who can say um you know roughly if these are OK and how far away they are.	Yeah.	Um but I think it 's gotta be pretty good because otherwise the word recognition would be really b crummy.	false	QMSum_225
Oh or i have someone look at the alignments uh maybe a linguist who can say um you know roughly if these are OK and how far away they are.	Yeah.	Um but I think it 's gotta be pretty good because otherwise the word recognition would be really b crummy.	Right , right.	false	QMSum_225
Yeah.	Um but I think it 's gotta be pretty good because otherwise the word recognition would be really b crummy.	Right , right.	It wouldn't necessarily be the other way around , if the wor word recognition was crummy the alignment might be OK but if the word recognition is this good the alignment should be pretty good. So that 's about it.	false	QMSum_225
Um but I think it 's gotta be pretty good because otherwise the word recognition would be really b crummy.	Right , right.	It wouldn't necessarily be the other way around , if the wor word recognition was crummy the alignment might be OK but if the word recognition is this good the alignment should be pretty good. So that 's about it.	I r	false	QMSum_225
Right , right.	It wouldn't necessarily be the other way around , if the wor word recognition was crummy the alignment might be OK but if the word recognition is this good the alignment should be pretty good. So that 's about it.	I r	I wonder if this is a good thing or a bad thing though , I mean if we 're pr	false	QMSum_225
It wouldn't necessarily be the other way around , if the wor word recognition was crummy the alignment might be OK but if the word recognition is this good the alignment should be pretty good. So that 's about it.	I r	I wonder if this is a good thing or a bad thing though , I mean if we 're pr	That we 're starting so well ?	false	QMSum_225
I r	I wonder if this is a good thing or a bad thing though , I mean if we 're pr	That we 're starting so well ?	yeah if we 're producing a database that everybody 's gonna do well on	false	QMSum_225
I wonder if this is a good thing or a bad thing though , I mean if we 're pr	That we 're starting so well ?	yeah if we 're producing a database that everybody 's gonna do well on	Oh	false	QMSum_225
That we 're starting so well ?	yeah if we 're producing a database that everybody 's gonna do well on	Oh	Don't worry about it w d that 's that 's the close talking mikes. Try it on the P Z Ms and and	false	QMSum_225
yeah if we 're producing a database that everybody 's gonna do well on	Oh	Don't worry about it w d that 's that 's the close talking mikes. Try it on the P Z Ms and and	Yeah , which I would which well n n n n	false	QMSum_225
Oh	Don't worry about it w d that 's that 's the close talking mikes. Try it on the P Z Ms and and	Yeah , which I would which well n n n n	So the real value of the database is these ?	false	QMSum_225
Don't worry about it w d that 's that 's the close talking mikes. Try it on the P Z Ms and and	Yeah , which I would which well n n n n	So the real value of the database is these ?	Yeah , yeah , yeah , yeah.	false	QMSum_225
Yeah , which I would which well n n n n	So the real value of the database is these ?	Yeah , yeah , yeah , yeah.	Yeah , abso well no but	false	QMSum_225
So the real value of the database is these ?	Yeah , yeah , yeah , yeah.	Yeah , abso well no but	I mean there 's still just the w the percentages and , I mean they 're not a as we 've talked about before there 's probably overlaps	false	QMSum_225
Yeah , yeah , yeah , yeah.	Yeah , abso well no but	I mean there 's still just the w the percentages and , I mean they 're not a as we 've talked about before there 's probably overlaps	This i yeah. This is not that good.	false	QMSum_225
Yeah , abso well no but	I mean there 's still just the w the percentages and , I mean they 're not a as we 've talked about before there 's probably overlaps	This i yeah. This is not that good.	there 's probably overlaps in in uh in fair number in Switchboard as well so but but there 's other phenomena , it 's a meeting , it 's a different thing and there 's lots of stuff to learn with the close talking mikes but uh yeah certainly I 'd like to see as soon as we could , I mean maybe get some of the glitches out of the way but soon as we could how well it does with say with the P Z Ms or maybe even one of the	false	QMSum_225
I mean there 's still just the w the percentages and , I mean they 're not a as we 've talked about before there 's probably overlaps	This i yeah. This is not that good.	there 's probably overlaps in in uh in fair number in Switchboard as well so but but there 's other phenomena , it 's a meeting , it 's a different thing and there 's lots of stuff to learn with the close talking mikes but uh yeah certainly I 'd like to see as soon as we could , I mean maybe get some of the glitches out of the way but soon as we could how well it does with say with the P Z Ms or maybe even one of the	Right.	false	QMSum_225
This i yeah. This is not that good.	there 's probably overlaps in in uh in fair number in Switchboard as well so but but there 's other phenomena , it 's a meeting , it 's a different thing and there 's lots of stuff to learn with the close talking mikes but uh yeah certainly I 'd like to see as soon as we could , I mean maybe get some of the glitches out of the way but soon as we could how well it does with say with the P Z Ms or maybe even one of the	Right.	and uh see if it 's , you know is it a hundred twenty percent or maybe it 's not maybe if with some adaptation you get this down to fifty percent or forty - five percent or something and and then if for the PZM it 's seventy or something like that that 's actually something we could sort of work with a little bit	false	QMSum_225
there 's probably overlaps in in uh in fair number in Switchboard as well so but but there 's other phenomena , it 's a meeting , it 's a different thing and there 's lots of stuff to learn with the close talking mikes but uh yeah certainly I 'd like to see as soon as we could , I mean maybe get some of the glitches out of the way but soon as we could how well it does with say with the P Z Ms or maybe even one of the	Right.	and uh see if it 's , you know is it a hundred twenty percent or maybe it 's not maybe if with some adaptation you get this down to fifty percent or forty - five percent or something and and then if for the PZM it 's seventy or something like that that 's actually something we could sort of work with a little bit	Yeah.	false	QMSum_225
Right.	and uh see if it 's , you know is it a hundred twenty percent or maybe it 's not maybe if with some adaptation you get this down to fifty percent or forty - five percent or something and and then if for the PZM it 's seventy or something like that that 's actually something we could sort of work with a little bit	Yeah.	so	true	QMSum_225
and uh see if it 's , you know is it a hundred twenty percent or maybe it 's not maybe if with some adaptation you get this down to fifty percent or forty - five percent or something and and then if for the PZM it 's seventy or something like that that 's actually something we could sort of work with a little bit	Yeah.	so	No I think it 's really , I mean this way we least have a baseline we know that for instance the transcripts are very good so once you can get to the words that the recognizer which is a total subset of the things you need to understand the the text um yeah they 're pretty good so and and it 's converting automatically from the XML to the chopping up the wave forms and so forth it 's not the case that the end of one utterance is in the next segment and things like that which we had more problems with in Switchboard so that 's good. And um let 's see there was one more thing I wanted to to mention I can't remember um Sorry can't remember. anyway it 's	false	QMSum_225
Yeah.	so	No I think it 's really , I mean this way we least have a baseline we know that for instance the transcripts are very good so once you can get to the words that the recognizer which is a total subset of the things you need to understand the the text um yeah they 're pretty good so and and it 's converting automatically from the XML to the chopping up the wave forms and so forth it 's not the case that the end of one utterance is in the next segment and things like that which we had more problems with in Switchboard so that 's good. And um let 's see there was one more thing I wanted to to mention I can't remember um Sorry can't remember. anyway it 's	Congratulations is really great.	false	QMSum_225
so	No I think it 's really , I mean this way we least have a baseline we know that for instance the transcripts are very good so once you can get to the words that the recognizer which is a total subset of the things you need to understand the the text um yeah they 're pretty good so and and it 's converting automatically from the XML to the chopping up the wave forms and so forth it 's not the case that the end of one utterance is in the next segment and things like that which we had more problems with in Switchboard so that 's good. And um let 's see there was one more thing I wanted to to mention I can't remember um Sorry can't remember. anyway it 's	Congratulations is really great.	Yeah.	false	QMSum_225
No I think it 's really , I mean this way we least have a baseline we know that for instance the transcripts are very good so once you can get to the words that the recognizer which is a total subset of the things you need to understand the the text um yeah they 're pretty good so and and it 's converting automatically from the XML to the chopping up the wave forms and so forth it 's not the case that the end of one utterance is in the next segment and things like that which we had more problems with in Switchboard so that 's good. And um let 's see there was one more thing I wanted to to mention I can't remember um Sorry can't remember. anyway it 's	Congratulations is really great.	Yeah.	well it was , I mean I really didn't do this myself	false	QMSum_225
Congratulations is really great.	Yeah.	well it was , I mean I really didn't do this myself	Yeah , it 's really good.	false	QMSum_225
Yeah.	well it was , I mean I really didn't do this myself	Yeah , it 's really good.	so Andreas set up this recognizer and by the way the recognizer all the files I 'm moving to SRI and running everything there so I brought back just these result files and people can look at them um so	false	QMSum_225
well it was , I mean I really didn't do this myself	Yeah , it 's really good.	so Andreas set up this recognizer and by the way the recognizer all the files I 'm moving to SRI and running everything there so I brought back just these result files and people can look at them um so	We we talked about setting up the SRI recognizer here. That 's you know if if there are more machines um uh here plus people can could run their own uh you know variants of of of the recognition runs um certainly doable. Um.	false	QMSum_225
Yeah , it 's really good.	so Andreas set up this recognizer and by the way the recognizer all the files I 'm moving to SRI and running everything there so I brought back just these result files and people can look at them um so	We we talked about setting up the SRI recognizer here. That 's you know if if there are more machines um uh here plus people can could run their own uh you know variants of of of the recognition runs um certainly doable. Um.	Yeah and well certainly if the recognition as opposed to training , yeah.	false	QMSum_225
so Andreas set up this recognizer and by the way the recognizer all the files I 'm moving to SRI and running everything there so I brought back just these result files and people can look at them um so	We we talked about setting up the SRI recognizer here. That 's you know if if there are more machines um uh here plus people can could run their own uh you know variants of of of the recognition runs um certainly doable. Um.	Yeah and well certainly if the recognition as opposed to training , yeah.	Yeah.	false	QMSum_225
We we talked about setting up the SRI recognizer here. That 's you know if if there are more machines um uh here plus people can could run their own uh you know variants of of of the recognition runs um certainly doable. Um.	Yeah and well certainly if the recognition as opposed to training , yeah.	Yeah.	Seems reasonable.	false	QMSum_225
Yeah and well certainly if the recognition as opposed to training , yeah.	Yeah.	Seems reasonable.	I need t Hmm. I need to ask one question.	false	QMSum_225
Yeah.	Seems reasonable.	I need t Hmm. I need to ask one question.	Yeah.	false	QMSum_225
Seems reasonable.	I need t Hmm. I need to ask one question.	Yeah.	Which is um so this issue of the uh legalistic aspects of the pre - sent you know pre - adapted Yeah , well , so what I mean is um the uh the data that you take into SRI , first first question , you 're maintaining it in in a place that wouldn't be publicly readable that that kind of stuff , right ?	false	QMSum_225
I need t Hmm. I need to ask one question.	Yeah.	Which is um so this issue of the uh legalistic aspects of the pre - sent you know pre - adapted Yeah , well , so what I mean is um the uh the data that you take into SRI , first first question , you 're maintaining it in in a place that wouldn't be publicly readable that that kind of stuff , right ?	U um	false	QMSum_225
Yeah.	Which is um so this issue of the uh legalistic aspects of the pre - sent you know pre - adapted Yeah , well , so what I mean is um the uh the data that you take into SRI , first first question , you 're maintaining it in in a place that wouldn't be publicly readable that that kind of stuff , right ?	U um	From the outside world or	false	QMSum_225
Which is um so this issue of the uh legalistic aspects of the pre - sent you know pre - adapted Yeah , well , so what I mean is um the uh the data that you take into SRI , first first question , you 're maintaining it in in a place that wouldn't be publicly readable that that kind of stuff , right ?	U um	From the outside world or	By uh people uh who are not associated with this project.	false	QMSum_225
U um	From the outside world or	By uh people uh who are not associated with this project.	Oh.	false	QMSum_225
From the outside world or	By uh people uh who are not associated with this project.	Oh.	It 's human subjects issues , I told you about that.	false	QMSum_225
By uh people uh who are not associated with this project.	Oh.	It 's human subjects issues , I told you about that.	Um oh.	false	QMSum_225
Oh.	It 's human subjects issues , I told you about that.	Um oh.	Exactly.	false	QMSum_225
It 's human subjects issues , I told you about that.	Um oh.	Exactly.	Well OK we have n no names. Although I sh um	false	QMSum_225
Um oh.	Exactly.	Well OK we have n no names. Although I sh um	That that 's not the issue ,	false	QMSum_225
Exactly.	Well OK we have n no names. Although I sh um	That that 's not the issue ,	de audio data itself ?	false	QMSum_225
Well OK we have n no names. Although I sh um	That that 's not the issue ,	de audio data itself ?	it 's just the audio data itself , until people have a chance to edit it.	false	QMSum_225
That that 's not the issue ,	de audio data itself ?	it 's just the audio data itself , until people have a chance to edit it.	Mm - hmm , exactly.	false	QMSum_225
de audio data itself ?	it 's just the audio data itself , until people have a chance to edit it.	Mm - hmm , exactly.	Uh so well I can I can protect my directories through there.	false	QMSum_225
it 's just the audio data itself , until people have a chance to edit it.	Mm - hmm , exactly.	Uh so well I can I can protect my directories through there.	Yeah.	false	QMSum_225
Mm - hmm , exactly.	Uh so well I can I can protect my directories through there.	Yeah.	Right now they 're not they 're in the speech group directories which so I will	false	QMSum_225
Uh so well I can I can protect my directories through there.	Yeah.	Right now they 're not they 're in the speech group directories which so I will	Great.	false	QMSum_225
Yeah.	Right now they 're not they 're in the speech group directories which so I will	Great.	I didn't know that actually.	false	QMSum_225
Right now they 're not they 're in the speech group directories which so I will	Great.	I didn't know that actually.	Yeah so we just have to go through this process of having people approve the transcriptions ,	false	QMSum_225
Great.	I didn't know that actually.	Yeah so we just have to go through this process of having people approve the transcriptions ,	Yeah OK.	false	QMSum_225
I didn't know that actually.	Yeah so we just have to go through this process of having people approve the transcriptions ,	Yeah OK.	say it 's OK.	false	QMSum_225
Yeah so we just have to go through this process of having people approve the transcriptions ,	Yeah OK.	say it 's OK.	Right OK.	false	QMSum_225
Yeah OK.	say it 's OK.	Right OK.	Yeah , we had to get them to approve em and then i cuz cuz the other question I was gonna ask is if we 're having um you know it 's but this this meeting that you have , no problem cuz I I well I mean I I speak for myself	false	QMSum_225
say it 's OK.	Right OK.	Yeah , we had to get them to approve em and then i cuz cuz the other question I was gonna ask is if we 're having um you know it 's but this this meeting that you have , no problem cuz I I well I mean I I speak for myself	It 's us.	false	QMSum_225
Right OK.	Yeah , we had to get them to approve em and then i cuz cuz the other question I was gonna ask is if we 're having um you know it 's but this this meeting that you have , no problem cuz I I well I mean I I speak for myself	It 's us.	but but I think that we didn't do anything that but well anyway so uh I wouldn't be too concerned about it with respect to that although we should clear it with Eric and Dan of course but these results are based on data which haven't had the uh haven't had the chance to be reviewed by the subjects	false	QMSum_225
Yeah , we had to get them to approve em and then i cuz cuz the other question I was gonna ask is if we 're having um you know it 's but this this meeting that you have , no problem cuz I I well I mean I I speak for myself	It 's us.	but but I think that we didn't do anything that but well anyway so uh I wouldn't be too concerned about it with respect to that although we should clear it with Eric and Dan of course but these results are based on data which haven't had the uh haven't had the chance to be reviewed by the subjects	That 's true.	false	QMSum_225
It 's us.	but but I think that we didn't do anything that but well anyway so uh I wouldn't be too concerned about it with respect to that although we should clear it with Eric and Dan of course but these results are based on data which haven't had the uh haven't had the chance to be reviewed by the subjects	That 's true.	and I don't know how that stands , I mean if you if you get fantastic results and it 's involving data which which later end up being lessened by , you know certain elisions , then I don't know but I wanted to raise that issue ,	false	QMSum_225
but but I think that we didn't do anything that but well anyway so uh I wouldn't be too concerned about it with respect to that although we should clear it with Eric and Dan of course but these results are based on data which haven't had the uh haven't had the chance to be reviewed by the subjects	That 's true.	and I don't know how that stands , I mean if you if you get fantastic results and it 's involving data which which later end up being lessened by , you know certain elisions , then I don't know but I wanted to raise that issue ,	Well we ,	false	QMSum_225
That 's true.	and I don't know how that stands , I mean if you if you get fantastic results and it 's involving data which which later end up being lessened by , you know certain elisions , then I don't know but I wanted to raise that issue ,	Well we ,	that 's all.	false	QMSum_225
and I don't know how that stands , I mean if you if you get fantastic results and it 's involving data which which later end up being lessened by , you know certain elisions , then I don't know but I wanted to raise that issue ,	Well we ,	that 's all.	I mean once we get all this streamlined it may be sh it hopefully it will be fairly quick but we get the transcriptions , people approve them and so on it 's just that we 're	false	QMSum_225
Well we ,	that 's all.	I mean once we get all this streamlined it may be sh it hopefully it will be fairly quick but we get the transcriptions , people approve them and so on it 's just that we 're	Alright we need to work at a system for doing that approval so that we can send people the transcripts	false	QMSum_225
that 's all.	I mean once we get all this streamlined it may be sh it hopefully it will be fairly quick but we get the transcriptions , people approve them and so on it 's just that we 're	Alright we need to work at a system for doing that approval so that we can send people the transcripts	Great.	false	QMSum_225
I mean once we get all this streamlined it may be sh it hopefully it will be fairly quick but we get the transcriptions , people approve them and so on it 's just that we 're	Alright we need to work at a system for doing that approval so that we can send people the transcripts	Great.	Mmm.	false	QMSum_225
Alright we need to work at a system for doing that approval so that we can send people the transcripts	Great.	Mmm.	Yeah.	false	QMSum_225
Great.	Mmm.	Yeah.	and get back any bleeps that they want	false	QMSum_225
Mmm.	Yeah.	and get back any bleeps that they want	Yeah actually the bleeps are also an issue I thought.	false	QMSum_225
Yeah.	and get back any bleeps that they want	Yeah actually the bleeps are also an issue I thought.	It 's gonna be a rare thing that there 's a bleep for the most part.	false	QMSum_225
and get back any bleeps that they want	Yeah actually the bleeps are also an issue I thought.	It 's gonna be a rare thing that there 's a bleep for the most part.	U uh actually I had a question about the downsampling , um I don't know who , I mean how this was done but is is there are there any um issues with downsampling	false	QMSum_225
Yeah actually the bleeps are also an issue I thought.	It 's gonna be a rare thing that there 's a bleep for the most part.	U uh actually I had a question about the downsampling , um I don't know who , I mean how this was done but is is there are there any um issues with downsampling	Don did this.	false	QMSum_225
It 's gonna be a rare thing that there 's a bleep for the most part.	U uh actually I had a question about the downsampling , um I don't know who , I mean how this was done but is is there are there any um issues with downsampling	Don did this.	because I know that the recognizer um that we use h can do it sort of on the fly um so we wouldn't have to have it eh you know do it uh explicitly beforehand. And is there any um i are there other d sev uh is there more than one way to do the downsampling where one might be better than another ?	false	QMSum_225
U uh actually I had a question about the downsampling , um I don't know who , I mean how this was done but is is there are there any um issues with downsampling	Don did this.	because I know that the recognizer um that we use h can do it sort of on the fly um so we wouldn't have to have it eh you know do it uh explicitly beforehand. And is there any um i are there other d sev uh is there more than one way to do the downsampling where one might be better than another ?	There are lots of w there are lots of ways to do the downsampling um different filters to put on ,	false	QMSum_225
Don did this.	because I know that the recognizer um that we use h can do it sort of on the fly um so we wouldn't have to have it eh you know do it uh explicitly beforehand. And is there any um i are there other d sev uh is there more than one way to do the downsampling where one might be better than another ?	There are lots of w there are lots of ways to do the downsampling um different filters to put on ,	OK. Right. OK.	false	QMSum_225
because I know that the recognizer um that we use h can do it sort of on the fly um so we wouldn't have to have it eh you know do it uh explicitly beforehand. And is there any um i are there other d sev uh is there more than one way to do the downsampling where one might be better than another ?	There are lots of w there are lots of ways to do the downsampling um different filters to put on ,	OK. Right. OK.	like anti - aliasing stuff.	false	QMSum_225
There are lots of w there are lots of ways to do the downsampling um different filters to put on ,	OK. Right. OK.	like anti - aliasing stuff.	So so the th	false	QMSum_225
OK. Right. OK.	like anti - aliasing stuff.	So so the th	I don't think we even know which one I assume you 're using syncat to do it ?	false	QMSum_225
like anti - aliasing stuff.	So so the th	I don't think we even know which one I assume you 're using syncat to do it ?	No , I 'm using uh SN SND uh are resample.	false	QMSum_225
So so the th	I don't think we even know which one I assume you 're using syncat to do it ?	No , I 'm using uh SN SND uh are resample.	Or sound resample ?	false	QMSum_225
I don't think we even know which one I assume you 're using syncat to do it ?	No , I 'm using uh SN SND uh are resample.	Or sound resample ?	Re - re ref	false	QMSum_225
No , I 'm using uh SN SND uh are resample.	Or sound resample ?	Re - re ref	Resample.	false	QMSum_225
Or sound resample ?	Re - re ref	Resample.	yeah.	false	QMSum_225
Re - re ref	Resample.	yeah.	Yeah and Dan 's archaic acronyms.	false	QMSum_225
Resample.	yeah.	Yeah and Dan 's archaic acronyms.	RSMP. Yeah , I don't really.	false	QMSum_225
yeah.	Yeah and Dan 's archaic acronyms.	RSMP. Yeah , I don't really.	Missing all the vowels.	false	QMSum_225
Yeah and Dan 's archaic acronyms.	RSMP. Yeah , I don't really.	Missing all the vowels.	I just yeah I found it.	false	QMSum_225
RSMP. Yeah , I don't really.	Missing all the vowels.	I just yeah I found it.	Some of the vowels ,	false	QMSum_225
Missing all the vowels.	I just yeah I found it.	Some of the vowels ,	Not all of them.	false	QMSum_225
I just yeah I found it.	Some of the vowels ,	Not all of them.	almost all the vowels , that 's the hard part.	false	QMSum_225
Some of the vowels ,	Not all of them.	almost all the vowels , that 's the hard part.	So so the other thing we should try is to just take the original wave forms ,	false	QMSum_225
Not all of them.	almost all the vowels , that 's the hard part.	So so the other thing we should try is to just take the original wave forms ,	And a few of the consonants.	false	QMSum_225
almost all the vowels , that 's the hard part.	So so the other thing we should try is to just take the original wave forms ,	And a few of the consonants.	I mean segment them but not downsample them.	false	QMSum_225

Well we 've done that for one meeting.

so. For forced alignment.

Uh oh oh f not for words

Ye - right.

false

QMSum_225

so. For forced alignment.

Uh oh oh f not for words

Ye - right.

I 'm sorry just for overlaps is we did it for not not for words.

false

QMSum_225

Uh oh oh f not for words

Ye - right.

I 'm sorry just for overlaps is we did it for not not for words.

Right. So this would be like if you take the words um you know and force align them on all the individual close talk uh close talking mikes then how good are these sort of in reality

false

QMSum_225

Ye - right.

I 'm sorry just for overlaps is we did it for not not for words.

Right. So this would be like if you take the words um you know and force align them on all the individual close talk uh close talking mikes then how good are these sort of in reality

Right.

false

QMSum_225

I 'm sorry just for overlaps is we did it for not not for words.

Right. So this would be like if you take the words um you know and force align them on all the individual close talk uh close talking mikes then how good are these sort of in reality

Right.

and then I was thinking it

false

QMSum_225

Right. So this would be like if you take the words um you know and force align them on all the individual close talk uh close talking mikes then how good are these sort of in reality

Right.

and then I was thinking it

So we might want to take twenty minutes and do a closer word level transcription. Maybe actually mark the word boundaries.

false

QMSum_225

Right.

and then I was thinking it

So we might want to take twenty minutes and do a closer word level transcription. Maybe actually mark the word boundaries.

Oh or i have someone look at the alignments uh maybe a linguist who can say um you know roughly if these are OK and how far away they are.

false

QMSum_225

and then I was thinking it

So we might want to take twenty minutes and do a closer word level transcription. Maybe actually mark the word boundaries.

Oh or i have someone look at the alignments uh maybe a linguist who can say um you know roughly if these are OK and how far away they are.

Yeah.

false

QMSum_225

So we might want to take twenty minutes and do a closer word level transcription. Maybe actually mark the word boundaries.

Oh or i have someone look at the alignments uh maybe a linguist who can say um you know roughly if these are OK and how far away they are.

Yeah.

Um but I think it 's gotta be pretty good because otherwise the word recognition would be really b crummy.

false

QMSum_225

Oh or i have someone look at the alignments uh maybe a linguist who can say um you know roughly if these are OK and how far away they are.

Yeah.

Um but I think it 's gotta be pretty good because otherwise the word recognition would be really b crummy.

Right , right.

false

QMSum_225

Yeah.

Um but I think it 's gotta be pretty good because otherwise the word recognition would be really b crummy.

Right , right.

It wouldn't necessarily be the other way around , if the wor word recognition was crummy the alignment might be OK but if the word recognition is this good the alignment should be pretty good. So that 's about it.

false

QMSum_225

Um but I think it 's gotta be pretty good because otherwise the word recognition would be really b crummy.

Right , right.

It wouldn't necessarily be the other way around , if the wor word recognition was crummy the alignment might be OK but if the word recognition is this good the alignment should be pretty good. So that 's about it.

I r

false

QMSum_225

Right , right.

It wouldn't necessarily be the other way around , if the wor word recognition was crummy the alignment might be OK but if the word recognition is this good the alignment should be pretty good. So that 's about it.

I r

I wonder if this is a good thing or a bad thing though , I mean if we 're pr

false

QMSum_225

It wouldn't necessarily be the other way around , if the wor word recognition was crummy the alignment might be OK but if the word recognition is this good the alignment should be pretty good. So that 's about it.

I r

I wonder if this is a good thing or a bad thing though , I mean if we 're pr

That we 're starting so well ?

false

QMSum_225

I r

I wonder if this is a good thing or a bad thing though , I mean if we 're pr

That we 're starting so well ?

yeah if we 're producing a database that everybody 's gonna do well on

false

QMSum_225

I wonder if this is a good thing or a bad thing though , I mean if we 're pr

That we 're starting so well ?

yeah if we 're producing a database that everybody 's gonna do well on

Oh

false

QMSum_225

That we 're starting so well ?

yeah if we 're producing a database that everybody 's gonna do well on

Oh

Don't worry about it w d that 's that 's the close talking mikes. Try it on the P Z Ms and and

false

QMSum_225

yeah if we 're producing a database that everybody 's gonna do well on

Oh

Don't worry about it w d that 's that 's the close talking mikes. Try it on the P Z Ms and and

Yeah , which I would which well n n n n

false

QMSum_225

Oh

Don't worry about it w d that 's that 's the close talking mikes. Try it on the P Z Ms and and

Yeah , which I would which well n n n n

So the real value of the database is these ?

false

QMSum_225

Don't worry about it w d that 's that 's the close talking mikes. Try it on the P Z Ms and and

Yeah , which I would which well n n n n

So the real value of the database is these ?

Yeah , yeah , yeah , yeah.

false

QMSum_225

Yeah , which I would which well n n n n

So the real value of the database is these ?

Yeah , yeah , yeah , yeah.

Yeah , abso well no but

false

QMSum_225

So the real value of the database is these ?

Yeah , yeah , yeah , yeah.

Yeah , abso well no but

I mean there 's still just the w the percentages and , I mean they 're not a as we 've talked about before there 's probably overlaps

false

QMSum_225

Yeah , yeah , yeah , yeah.

Yeah , abso well no but

I mean there 's still just the w the percentages and , I mean they 're not a as we 've talked about before there 's probably overlaps

This i yeah. This is not that good.

false

QMSum_225

Yeah , abso well no but

I mean there 's still just the w the percentages and , I mean they 're not a as we 've talked about before there 's probably overlaps

This i yeah. This is not that good.

there 's probably overlaps in in uh in fair number in Switchboard as well so but but there 's other phenomena , it 's a meeting , it 's a different thing and there 's lots of stuff to learn with the close talking mikes but uh yeah certainly I 'd like to see as soon as we could , I mean maybe get some of the glitches out of the way but soon as we could how well it does with say with the P Z Ms or maybe even one of the

false

QMSum_225

I mean there 's still just the w the percentages and , I mean they 're not a as we 've talked about before there 's probably overlaps

This i yeah. This is not that good.

there 's probably overlaps in in uh in fair number in Switchboard as well so but but there 's other phenomena , it 's a meeting , it 's a different thing and there 's lots of stuff to learn with the close talking mikes but uh yeah certainly I 'd like to see as soon as we could , I mean maybe get some of the glitches out of the way but soon as we could how well it does with say with the P Z Ms or maybe even one of the

Right.

false

QMSum_225

This i yeah. This is not that good.

there 's probably overlaps in in uh in fair number in Switchboard as well so but but there 's other phenomena , it 's a meeting , it 's a different thing and there 's lots of stuff to learn with the close talking mikes but uh yeah certainly I 'd like to see as soon as we could , I mean maybe get some of the glitches out of the way but soon as we could how well it does with say with the P Z Ms or maybe even one of the

Right.

and uh see if it 's , you know is it a hundred twenty percent or maybe it 's not maybe if with some adaptation you get this down to fifty percent or forty - five percent or something and and then if for the PZM it 's seventy or something like that that 's actually something we could sort of work with a little bit

false

QMSum_225

there 's probably overlaps in in uh in fair number in Switchboard as well so but but there 's other phenomena , it 's a meeting , it 's a different thing and there 's lots of stuff to learn with the close talking mikes but uh yeah certainly I 'd like to see as soon as we could , I mean maybe get some of the glitches out of the way but soon as we could how well it does with say with the P Z Ms or maybe even one of the

Right.

and uh see if it 's , you know is it a hundred twenty percent or maybe it 's not maybe if with some adaptation you get this down to fifty percent or forty - five percent or something and and then if for the PZM it 's seventy or something like that that 's actually something we could sort of work with a little bit

Yeah.

false

QMSum_225

Right.

and uh see if it 's , you know is it a hundred twenty percent or maybe it 's not maybe if with some adaptation you get this down to fifty percent or forty - five percent or something and and then if for the PZM it 's seventy or something like that that 's actually something we could sort of work with a little bit

Yeah.

so

true

QMSum_225

and uh see if it 's , you know is it a hundred twenty percent or maybe it 's not maybe if with some adaptation you get this down to fifty percent or forty - five percent or something and and then if for the PZM it 's seventy or something like that that 's actually something we could sort of work with a little bit

Yeah.

so

No I think it 's really , I mean this way we least have a baseline we know that for instance the transcripts are very good so once you can get to the words that the recognizer which is a total subset of the things you need to understand the the text um yeah they 're pretty good so and and it 's converting automatically from the XML to the chopping up the wave forms and so forth it 's not the case that the end of one utterance is in the next segment and things like that which we had more problems with in Switchboard so that 's good. And um let 's see there was one more thing I wanted to to mention I can't remember um Sorry can't remember. anyway it 's

false

QMSum_225

Yeah.

so

No I think it 's really , I mean this way we least have a baseline we know that for instance the transcripts are very good so once you can get to the words that the recognizer which is a total subset of the things you need to understand the the text um yeah they 're pretty good so and and it 's converting automatically from the XML to the chopping up the wave forms and so forth it 's not the case that the end of one utterance is in the next segment and things like that which we had more problems with in Switchboard so that 's good. And um let 's see there was one more thing I wanted to to mention I can't remember um Sorry can't remember. anyway it 's

Congratulations is really great.

false

QMSum_225

so

No I think it 's really , I mean this way we least have a baseline we know that for instance the transcripts are very good so once you can get to the words that the recognizer which is a total subset of the things you need to understand the the text um yeah they 're pretty good so and and it 's converting automatically from the XML to the chopping up the wave forms and so forth it 's not the case that the end of one utterance is in the next segment and things like that which we had more problems with in Switchboard so that 's good. And um let 's see there was one more thing I wanted to to mention I can't remember um Sorry can't remember. anyway it 's

Congratulations is really great.

Yeah.

false

QMSum_225

No I think it 's really , I mean this way we least have a baseline we know that for instance the transcripts are very good so once you can get to the words that the recognizer which is a total subset of the things you need to understand the the text um yeah they 're pretty good so and and it 's converting automatically from the XML to the chopping up the wave forms and so forth it 's not the case that the end of one utterance is in the next segment and things like that which we had more problems with in Switchboard so that 's good. And um let 's see there was one more thing I wanted to to mention I can't remember um Sorry can't remember. anyway it 's

Congratulations is really great.

Yeah.

well it was , I mean I really didn't do this myself

false

QMSum_225

Congratulations is really great.

Yeah.

well it was , I mean I really didn't do this myself

Yeah , it 's really good.

false

QMSum_225

Yeah.

well it was , I mean I really didn't do this myself

Yeah , it 's really good.

so Andreas set up this recognizer and by the way the recognizer all the files I 'm moving to SRI and running everything there so I brought back just these result files and people can look at them um so

false

QMSum_225

well it was , I mean I really didn't do this myself

Yeah , it 's really good.

so Andreas set up this recognizer and by the way the recognizer all the files I 'm moving to SRI and running everything there so I brought back just these result files and people can look at them um so

We we talked about setting up the SRI recognizer here. That 's you know if if there are more machines um uh here plus people can could run their own uh you know variants of of of the recognition runs um certainly doable. Um.

false

QMSum_225

Yeah , it 's really good.

so Andreas set up this recognizer and by the way the recognizer all the files I 'm moving to SRI and running everything there so I brought back just these result files and people can look at them um so

We we talked about setting up the SRI recognizer here. That 's you know if if there are more machines um uh here plus people can could run their own uh you know variants of of of the recognition runs um certainly doable. Um.

Yeah and well certainly if the recognition as opposed to training , yeah.

false

QMSum_225

so Andreas set up this recognizer and by the way the recognizer all the files I 'm moving to SRI and running everything there so I brought back just these result files and people can look at them um so

We we talked about setting up the SRI recognizer here. That 's you know if if there are more machines um uh here plus people can could run their own uh you know variants of of of the recognition runs um certainly doable. Um.

Yeah and well certainly if the recognition as opposed to training , yeah.

Yeah.

false

QMSum_225

We we talked about setting up the SRI recognizer here. That 's you know if if there are more machines um uh here plus people can could run their own uh you know variants of of of the recognition runs um certainly doable. Um.

Yeah and well certainly if the recognition as opposed to training , yeah.

Yeah.

Seems reasonable.

false

QMSum_225

Yeah and well certainly if the recognition as opposed to training , yeah.

Yeah.

Seems reasonable.

I need t Hmm. I need to ask one question.

false

QMSum_225

Yeah.

Seems reasonable.

I need t Hmm. I need to ask one question.

Yeah.

false

QMSum_225

Seems reasonable.

I need t Hmm. I need to ask one question.

Yeah.

Which is um so this issue of the uh legalistic aspects of the pre - sent you know pre - adapted Yeah , well , so what I mean is um the uh the data that you take into SRI , first first question , you 're maintaining it in in a place that wouldn't be publicly readable that that kind of stuff , right ?

false

QMSum_225

I need t Hmm. I need to ask one question.

Yeah.

Which is um so this issue of the uh legalistic aspects of the pre - sent you know pre - adapted Yeah , well , so what I mean is um the uh the data that you take into SRI , first first question , you 're maintaining it in in a place that wouldn't be publicly readable that that kind of stuff , right ?

U um

false

QMSum_225

Yeah.

Which is um so this issue of the uh legalistic aspects of the pre - sent you know pre - adapted Yeah , well , so what I mean is um the uh the data that you take into SRI , first first question , you 're maintaining it in in a place that wouldn't be publicly readable that that kind of stuff , right ?

U um

From the outside world or

false

QMSum_225

Which is um so this issue of the uh legalistic aspects of the pre - sent you know pre - adapted Yeah , well , so what I mean is um the uh the data that you take into SRI , first first question , you 're maintaining it in in a place that wouldn't be publicly readable that that kind of stuff , right ?

U um

From the outside world or

By uh people uh who are not associated with this project.

false

QMSum_225

U um

From the outside world or

By uh people uh who are not associated with this project.

Oh.

false

QMSum_225

From the outside world or

By uh people uh who are not associated with this project.

Oh.

It 's human subjects issues , I told you about that.

false

QMSum_225

By uh people uh who are not associated with this project.

Oh.

It 's human subjects issues , I told you about that.

Um oh.

false

QMSum_225

Oh.

It 's human subjects issues , I told you about that.

Um oh.

Exactly.

false

QMSum_225

It 's human subjects issues , I told you about that.

Um oh.

Exactly.

Well OK we have n no names. Although I sh um

false

QMSum_225

Um oh.

Exactly.

Well OK we have n no names. Although I sh um

That that 's not the issue ,

false

QMSum_225

Exactly.

Well OK we have n no names. Although I sh um

That that 's not the issue ,

de audio data itself ?

false

QMSum_225

Well OK we have n no names. Although I sh um

That that 's not the issue ,

de audio data itself ?

it 's just the audio data itself , until people have a chance to edit it.

false

QMSum_225

That that 's not the issue ,

de audio data itself ?

it 's just the audio data itself , until people have a chance to edit it.

Mm - hmm , exactly.

false

QMSum_225

de audio data itself ?

it 's just the audio data itself , until people have a chance to edit it.

Mm - hmm , exactly.

Uh so well I can I can protect my directories through there.

false

QMSum_225

it 's just the audio data itself , until people have a chance to edit it.

Mm - hmm , exactly.

Uh so well I can I can protect my directories through there.

Yeah.

false

QMSum_225

Mm - hmm , exactly.

Uh so well I can I can protect my directories through there.

Yeah.

Right now they 're not they 're in the speech group directories which so I will

false

QMSum_225

Uh so well I can I can protect my directories through there.

Yeah.

Right now they 're not they 're in the speech group directories which so I will

Great.

false

QMSum_225

Yeah.

Right now they 're not they 're in the speech group directories which so I will

Great.

I didn't know that actually.

false

QMSum_225

Right now they 're not they 're in the speech group directories which so I will

Great.

I didn't know that actually.

Yeah so we just have to go through this process of having people approve the transcriptions ,

false

QMSum_225

Great.

I didn't know that actually.

Yeah so we just have to go through this process of having people approve the transcriptions ,

Yeah OK.

false

QMSum_225

I didn't know that actually.

Yeah so we just have to go through this process of having people approve the transcriptions ,

Yeah OK.

say it 's OK.

false

QMSum_225

Yeah so we just have to go through this process of having people approve the transcriptions ,

Yeah OK.

say it 's OK.

Right OK.

false

QMSum_225

Yeah OK.

say it 's OK.

Right OK.

Yeah , we had to get them to approve em and then i cuz cuz the other question I was gonna ask is if we 're having um you know it 's but this this meeting that you have , no problem cuz I I well I mean I I speak for myself

false

QMSum_225

say it 's OK.

Right OK.

Yeah , we had to get them to approve em and then i cuz cuz the other question I was gonna ask is if we 're having um you know it 's but this this meeting that you have , no problem cuz I I well I mean I I speak for myself

It 's us.

false

QMSum_225

Right OK.

Yeah , we had to get them to approve em and then i cuz cuz the other question I was gonna ask is if we 're having um you know it 's but this this meeting that you have , no problem cuz I I well I mean I I speak for myself

It 's us.

but but I think that we didn't do anything that but well anyway so uh I wouldn't be too concerned about it with respect to that although we should clear it with Eric and Dan of course but these results are based on data which haven't had the uh haven't had the chance to be reviewed by the subjects

false

QMSum_225

Yeah , we had to get them to approve em and then i cuz cuz the other question I was gonna ask is if we 're having um you know it 's but this this meeting that you have , no problem cuz I I well I mean I I speak for myself

It 's us.

but but I think that we didn't do anything that but well anyway so uh I wouldn't be too concerned about it with respect to that although we should clear it with Eric and Dan of course but these results are based on data which haven't had the uh haven't had the chance to be reviewed by the subjects

That 's true.

false

QMSum_225

It 's us.

but but I think that we didn't do anything that but well anyway so uh I wouldn't be too concerned about it with respect to that although we should clear it with Eric and Dan of course but these results are based on data which haven't had the uh haven't had the chance to be reviewed by the subjects

That 's true.

and I don't know how that stands , I mean if you if you get fantastic results and it 's involving data which which later end up being lessened by , you know certain elisions , then I don't know but I wanted to raise that issue ,

false

QMSum_225

but but I think that we didn't do anything that but well anyway so uh I wouldn't be too concerned about it with respect to that although we should clear it with Eric and Dan of course but these results are based on data which haven't had the uh haven't had the chance to be reviewed by the subjects

That 's true.

and I don't know how that stands , I mean if you if you get fantastic results and it 's involving data which which later end up being lessened by , you know certain elisions , then I don't know but I wanted to raise that issue ,

Well we ,

false

QMSum_225

That 's true.

and I don't know how that stands , I mean if you if you get fantastic results and it 's involving data which which later end up being lessened by , you know certain elisions , then I don't know but I wanted to raise that issue ,

Well we ,

that 's all.

false

QMSum_225

and I don't know how that stands , I mean if you if you get fantastic results and it 's involving data which which later end up being lessened by , you know certain elisions , then I don't know but I wanted to raise that issue ,

Well we ,

that 's all.

I mean once we get all this streamlined it may be sh it hopefully it will be fairly quick but we get the transcriptions , people approve them and so on it 's just that we 're

false

QMSum_225

Well we ,

that 's all.

I mean once we get all this streamlined it may be sh it hopefully it will be fairly quick but we get the transcriptions , people approve them and so on it 's just that we 're

Alright we need to work at a system for doing that approval so that we can send people the transcripts

false

QMSum_225

that 's all.

I mean once we get all this streamlined it may be sh it hopefully it will be fairly quick but we get the transcriptions , people approve them and so on it 's just that we 're

Alright we need to work at a system for doing that approval so that we can send people the transcripts

Great.

false

QMSum_225

I mean once we get all this streamlined it may be sh it hopefully it will be fairly quick but we get the transcriptions , people approve them and so on it 's just that we 're

Alright we need to work at a system for doing that approval so that we can send people the transcripts

Great.

Mmm.

false

QMSum_225

Alright we need to work at a system for doing that approval so that we can send people the transcripts

Great.

Mmm.

Yeah.

false

QMSum_225

Great.

Mmm.

Yeah.

and get back any bleeps that they want

false

QMSum_225

Mmm.

Yeah.

and get back any bleeps that they want

Yeah actually the bleeps are also an issue I thought.

false

QMSum_225

Yeah.

and get back any bleeps that they want

Yeah actually the bleeps are also an issue I thought.

It 's gonna be a rare thing that there 's a bleep for the most part.

false

QMSum_225

and get back any bleeps that they want

Yeah actually the bleeps are also an issue I thought.

It 's gonna be a rare thing that there 's a bleep for the most part.

U uh actually I had a question about the downsampling , um I don't know who , I mean how this was done but is is there are there any um issues with downsampling

false

QMSum_225

Yeah actually the bleeps are also an issue I thought.

It 's gonna be a rare thing that there 's a bleep for the most part.

U uh actually I had a question about the downsampling , um I don't know who , I mean how this was done but is is there are there any um issues with downsampling

Don did this.

false

QMSum_225

It 's gonna be a rare thing that there 's a bleep for the most part.

U uh actually I had a question about the downsampling , um I don't know who , I mean how this was done but is is there are there any um issues with downsampling

Don did this.

because I know that the recognizer um that we use h can do it sort of on the fly um so we wouldn't have to have it eh you know do it uh explicitly beforehand. And is there any um i are there other d sev uh is there more than one way to do the downsampling where one might be better than another ?

false

QMSum_225

U uh actually I had a question about the downsampling , um I don't know who , I mean how this was done but is is there are there any um issues with downsampling

Don did this.

because I know that the recognizer um that we use h can do it sort of on the fly um so we wouldn't have to have it eh you know do it uh explicitly beforehand. And is there any um i are there other d sev uh is there more than one way to do the downsampling where one might be better than another ?

There are lots of w there are lots of ways to do the downsampling um different filters to put on ,

false

QMSum_225

Don did this.

because I know that the recognizer um that we use h can do it sort of on the fly um so we wouldn't have to have it eh you know do it uh explicitly beforehand. And is there any um i are there other d sev uh is there more than one way to do the downsampling where one might be better than another ?

There are lots of w there are lots of ways to do the downsampling um different filters to put on ,

OK. Right. OK.

false

QMSum_225

because I know that the recognizer um that we use h can do it sort of on the fly um so we wouldn't have to have it eh you know do it uh explicitly beforehand. And is there any um i are there other d sev uh is there more than one way to do the downsampling where one might be better than another ?

There are lots of w there are lots of ways to do the downsampling um different filters to put on ,

OK. Right. OK.

like anti - aliasing stuff.

false

QMSum_225

There are lots of w there are lots of ways to do the downsampling um different filters to put on ,

OK. Right. OK.

like anti - aliasing stuff.

So so the th

false

QMSum_225

OK. Right. OK.

like anti - aliasing stuff.

So so the th

I don't think we even know which one I assume you 're using syncat to do it ?

false

QMSum_225

like anti - aliasing stuff.

So so the th

I don't think we even know which one I assume you 're using syncat to do it ?

No , I 'm using uh SN SND uh are resample.

false

QMSum_225

So so the th

I don't think we even know which one I assume you 're using syncat to do it ?

No , I 'm using uh SN SND uh are resample.

Or sound resample ?

false

QMSum_225

I don't think we even know which one I assume you 're using syncat to do it ?

No , I 'm using uh SN SND uh are resample.

Or sound resample ?

Re - re ref

false

QMSum_225

No , I 'm using uh SN SND uh are resample.

Or sound resample ?

Re - re ref

Resample.

false

QMSum_225

Or sound resample ?

Re - re ref

Resample.

yeah.

false

QMSum_225

Re - re ref

Resample.

yeah.

Yeah and Dan 's archaic acronyms.

false

QMSum_225

Resample.

yeah.

Yeah and Dan 's archaic acronyms.

RSMP. Yeah , I don't really.

false

QMSum_225

yeah.

Yeah and Dan 's archaic acronyms.

RSMP. Yeah , I don't really.

Missing all the vowels.

false

QMSum_225

Yeah and Dan 's archaic acronyms.

RSMP. Yeah , I don't really.

Missing all the vowels.

I just yeah I found it.

false

QMSum_225

RSMP. Yeah , I don't really.

Missing all the vowels.

I just yeah I found it.

Some of the vowels ,

false

QMSum_225

Missing all the vowels.

I just yeah I found it.

Some of the vowels ,

Not all of them.

false

QMSum_225

I just yeah I found it.

Some of the vowels ,

Not all of them.

almost all the vowels , that 's the hard part.

false

QMSum_225

Some of the vowels ,

Not all of them.

almost all the vowels , that 's the hard part.

So so the other thing we should try is to just take the original wave forms ,

false

QMSum_225

Not all of them.

almost all the vowels , that 's the hard part.

So so the other thing we should try is to just take the original wave forms ,

And a few of the consonants.

false

QMSum_225

almost all the vowels , that 's the hard part.

So so the other thing we should try is to just take the original wave forms ,

And a few of the consonants.

I mean segment them but not downsample them.

false

QMSum_225