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
|
|---|---|---|---|---|---|
Um , we just How did we do it up again ?
|
Uh , we distend the we have the twenty - three coefficient af after the mel f filter ,
|
Mm - hmm.
|
and we extend these coefficient between the all the frequency range.
| false |
QMSum_120
|
Uh , we distend the we have the twenty - three coefficient af after the mel f filter ,
|
Mm - hmm.
|
and we extend these coefficient between the all the frequency range.
|
Mm - hmm.
| false |
QMSum_120
|
Mm - hmm.
|
and we extend these coefficient between the all the frequency range.
|
Mm - hmm.
|
And i the interpolation i between the point is give for the triang triangular filter , the value of the triangular filter and of this way we obtained this mode this model speech.
| false |
QMSum_120
|
and we extend these coefficient between the all the frequency range.
|
Mm - hmm.
|
And i the interpolation i between the point is give for the triang triangular filter , the value of the triangular filter and of this way we obtained this mode this model speech.
|
So you essentially take the values that th that you get from the triangular filter and extend them to sor sort of like a rectangle , that 's at that m value.
| false |
QMSum_120
|
Mm - hmm.
|
And i the interpolation i between the point is give for the triang triangular filter , the value of the triangular filter and of this way we obtained this mode this model speech.
|
So you essentially take the values that th that you get from the triangular filter and extend them to sor sort of like a rectangle , that 's at that m value.
|
Yeah.
| false |
QMSum_120
|
And i the interpolation i between the point is give for the triang triangular filter , the value of the triangular filter and of this way we obtained this mode this model speech.
|
So you essentially take the values that th that you get from the triangular filter and extend them to sor sort of like a rectangle , that 's at that m value.
|
Yeah.
|
Yeah. I think we have linear interpolation.
| false |
QMSum_120
|
So you essentially take the values that th that you get from the triangular filter and extend them to sor sort of like a rectangle , that 's at that m value.
|
Yeah.
|
Yeah. I think we have linear interpolation.
|
Mm - hmm.
| false |
QMSum_120
|
Yeah.
|
Yeah. I think we have linear interpolation.
|
Mm - hmm.
|
So we have we have one point for one energy for each filter bank ,
| false |
QMSum_120
|
Yeah. I think we have linear interpolation.
|
Mm - hmm.
|
So we have we have one point for one energy for each filter bank ,
|
mmm Yeah , it 's linear.
| false |
QMSum_120
|
Mm - hmm.
|
So we have we have one point for one energy for each filter bank ,
|
mmm Yeah , it 's linear.
|
Mmm.
| false |
QMSum_120
|
So we have we have one point for one energy for each filter bank ,
|
mmm Yeah , it 's linear.
|
Mmm.
|
Oh.
| false |
QMSum_120
|
mmm Yeah , it 's linear.
|
Mmm.
|
Oh.
|
which is the energy that 's centered on on on the triangle
| false |
QMSum_120
|
Mmm.
|
Oh.
|
which is the energy that 's centered on on on the triangle
|
Yeah. At the n at the center of the filter
| false |
QMSum_120
|
Oh.
|
which is the energy that 's centered on on on the triangle
|
Yeah. At the n at the center of the filter
|
So you you end up with a vector that 's the same length as the FFT vector ?
| false |
QMSum_120
|
which is the energy that 's centered on on on the triangle
|
Yeah. At the n at the center of the filter
|
So you you end up with a vector that 's the same length as the FFT vector ?
|
Yeah. That 's right.
| false |
QMSum_120
|
Yeah. At the n at the center of the filter
|
So you you end up with a vector that 's the same length as the FFT vector ?
|
Yeah. That 's right.
|
Yeah.
| false |
QMSum_120
|
So you you end up with a vector that 's the same length as the FFT vector ?
|
Yeah. That 's right.
|
Yeah.
|
And then you just , uh , compute differences
| false |
QMSum_120
|
Yeah. That 's right.
|
Yeah.
|
And then you just , uh , compute differences
|
Yeah. I have here one example if you if you want see something like that.
| false |
QMSum_120
|
Yeah.
|
And then you just , uh , compute differences
|
Yeah. I have here one example if you if you want see something like that.
|
Then we compute the difference.
| false |
QMSum_120
|
And then you just , uh , compute differences
|
Yeah. I have here one example if you if you want see something like that.
|
Then we compute the difference.
|
and ,
| false |
QMSum_120
|
Yeah. I have here one example if you if you want see something like that.
|
Then we compute the difference.
|
and ,
|
Yeah. Uh - huh.
| false |
QMSum_120
|
Then we compute the difference.
|
and ,
|
Yeah. Uh - huh.
|
OK.
| false |
QMSum_120
|
and ,
|
Yeah. Uh - huh.
|
OK.
|
uh , sum the differences ?
| false |
QMSum_120
|
Yeah. Uh - huh.
|
OK.
|
uh , sum the differences ?
|
So. And I think the variance is computed only from , like , two hundred hertz to one to fifteen hundred.
| false |
QMSum_120
|
OK.
|
uh , sum the differences ?
|
So. And I think the variance is computed only from , like , two hundred hertz to one to fifteen hundred.
|
Oh ! OK.
| false |
QMSum_120
|
uh , sum the differences ?
|
So. And I think the variance is computed only from , like , two hundred hertz to one to fifteen hundred.
|
Oh ! OK.
|
Mm - hmm.
| false |
QMSum_120
|
So. And I think the variance is computed only from , like , two hundred hertz to one to fifteen hundred.
|
Oh ! OK.
|
Mm - hmm.
|
Two thou two fifteen hundred ?
| false |
QMSum_120
|
Oh ! OK.
|
Mm - hmm.
|
Two thou two fifteen hundred ?
|
Mm - hmm.
| false |
QMSum_120
|
Mm - hmm.
|
Two thou two fifteen hundred ?
|
Mm - hmm.
|
Because
| false |
QMSum_120
|
Two thou two fifteen hundred ?
|
Mm - hmm.
|
Because
|
No.
| false |
QMSum_120
|
Mm - hmm.
|
Because
|
No.
|
Right.
| false |
QMSum_120
|
Because
|
No.
|
Right.
|
Two hundred and fifty thousand.
| false |
QMSum_120
|
No.
|
Right.
|
Two hundred and fifty thousand.
|
Fifteen hundred. Because Yeah.
| false |
QMSum_120
|
Right.
|
Two hundred and fifty thousand.
|
Fifteen hundred. Because Yeah.
|
Yeah. Two thousand and fifteen hundred.
| false |
QMSum_120
|
Two hundred and fifty thousand.
|
Fifteen hundred. Because Yeah.
|
Yeah. Two thousand and fifteen hundred.
|
Above , um it seems that Well , some voiced sound can have also , like , a noisy part on high frequencies , and But
| false |
QMSum_120
|
Fifteen hundred. Because Yeah.
|
Yeah. Two thousand and fifteen hundred.
|
Above , um it seems that Well , some voiced sound can have also , like , a noisy part on high frequencies , and But
|
Yeah.
| false |
QMSum_120
|
Yeah. Two thousand and fifteen hundred.
|
Above , um it seems that Well , some voiced sound can have also , like , a noisy part on high frequencies , and But
|
Yeah.
|
Well , it 's just
| false |
QMSum_120
|
Above , um it seems that Well , some voiced sound can have also , like , a noisy part on high frequencies , and But
|
Yeah.
|
Well , it 's just
|
No , it 's makes sense to look at low frequencies.
| false |
QMSum_120
|
Yeah.
|
Well , it 's just
|
No , it 's makes sense to look at low frequencies.
|
So this is uh , basically this is comparing an original version of the signal to a smoothed version of the same signal ?
| false |
QMSum_120
|
Well , it 's just
|
No , it 's makes sense to look at low frequencies.
|
So this is uh , basically this is comparing an original version of the signal to a smoothed version of the same signal ?
|
Yeah.
| false |
QMSum_120
|
No , it 's makes sense to look at low frequencies.
|
So this is uh , basically this is comparing an original version of the signal to a smoothed version of the same signal ?
|
Yeah.
|
Right. So i so i i this is I mean , i you could argue about whether it should be linear interpolation or or or or zeroeth order , but but
| false |
QMSum_120
|
So this is uh , basically this is comparing an original version of the signal to a smoothed version of the same signal ?
|
Yeah.
|
Right. So i so i i this is I mean , i you could argue about whether it should be linear interpolation or or or or zeroeth order , but but
|
Uh - huh.
| false |
QMSum_120
|
Yeah.
|
Right. So i so i i this is I mean , i you could argue about whether it should be linear interpolation or or or or zeroeth order , but but
|
Uh - huh.
|
at any rate something like this is what you 're feeding your recognizer , typically.
| false |
QMSum_120
|
Right. So i so i i this is I mean , i you could argue about whether it should be linear interpolation or or or or zeroeth order , but but
|
Uh - huh.
|
at any rate something like this is what you 're feeding your recognizer , typically.
|
Like which of the ?
| false |
QMSum_120
|
Uh - huh.
|
at any rate something like this is what you 're feeding your recognizer , typically.
|
Like which of the ?
|
No. Uh , so the mel cepstrum is the is the is the cepstrum of this this , uh , spectrum or log spectrum ,
| false |
QMSum_120
|
at any rate something like this is what you 're feeding your recognizer , typically.
|
Like which of the ?
|
No. Uh , so the mel cepstrum is the is the is the cepstrum of this this , uh , spectrum or log spectrum ,
|
So this is Yeah.
| false |
QMSum_120
|
Like which of the ?
|
No. Uh , so the mel cepstrum is the is the is the cepstrum of this this , uh , spectrum or log spectrum ,
|
So this is Yeah.
|
Yeah. Right , right.
| false |
QMSum_120
|
No. Uh , so the mel cepstrum is the is the is the cepstrum of this this , uh , spectrum or log spectrum ,
|
So this is Yeah.
|
Yeah. Right , right.
|
whatever it You - you 're subtracting in in in power domain or log domain ?
| false |
QMSum_120
|
So this is Yeah.
|
Yeah. Right , right.
|
whatever it You - you 're subtracting in in in power domain or log domain ?
|
In log domain. Yeah.
| false |
QMSum_120
|
Yeah. Right , right.
|
whatever it You - you 're subtracting in in in power domain or log domain ?
|
In log domain. Yeah.
|
Log domain.
| false |
QMSum_120
|
whatever it You - you 're subtracting in in in power domain or log domain ?
|
In log domain. Yeah.
|
Log domain.
|
OK. So it 's sort of like division , when you do the yeah , the spectra.
| false |
QMSum_120
|
In log domain. Yeah.
|
Log domain.
|
OK. So it 's sort of like division , when you do the yeah , the spectra.
|
Yeah.
| false |
QMSum_120
|
Log domain.
|
OK. So it 's sort of like division , when you do the yeah , the spectra.
|
Yeah.
|
Uh , yeah.
| false |
QMSum_120
|
OK. So it 's sort of like division , when you do the yeah , the spectra.
|
Yeah.
|
Uh , yeah.
|
It 's the ratio.
| false |
QMSum_120
|
Yeah.
|
Uh , yeah.
|
It 's the ratio.
|
Um. Yeah. But , anyway , um and that 's
| false |
QMSum_120
|
Uh , yeah.
|
It 's the ratio.
|
Um. Yeah. But , anyway , um and that 's
|
So what 's th uh , what 's the intuition behind this kind of a thing ? I I don't know really know the signal - processing well enough to understand what what is that doing.
| false |
QMSum_120
|
It 's the ratio.
|
Um. Yeah. But , anyway , um and that 's
|
So what 's th uh , what 's the intuition behind this kind of a thing ? I I don't know really know the signal - processing well enough to understand what what is that doing.
|
So. Yeah. What happen if what we have have what we would like to have is some spectrum of the excitation signal ,
| false |
QMSum_120
|
Um. Yeah. But , anyway , um and that 's
|
So what 's th uh , what 's the intuition behind this kind of a thing ? I I don't know really know the signal - processing well enough to understand what what is that doing.
|
So. Yeah. What happen if what we have have what we would like to have is some spectrum of the excitation signal ,
|
Yeah. I guess that makes sense. Yeah.
| false |
QMSum_120
|
So what 's th uh , what 's the intuition behind this kind of a thing ? I I don't know really know the signal - processing well enough to understand what what is that doing.
|
So. Yeah. What happen if what we have have what we would like to have is some spectrum of the excitation signal ,
|
Yeah. I guess that makes sense. Yeah.
|
which is for voiced sound ideally a a pulse train
| false |
QMSum_120
|
So. Yeah. What happen if what we have have what we would like to have is some spectrum of the excitation signal ,
|
Yeah. I guess that makes sense. Yeah.
|
which is for voiced sound ideally a a pulse train
|
Uh - huh.
| false |
QMSum_120
|
Yeah. I guess that makes sense. Yeah.
|
which is for voiced sound ideally a a pulse train
|
Uh - huh.
|
and for unvoiced it 's something that 's more flat.
| false |
QMSum_120
|
which is for voiced sound ideally a a pulse train
|
Uh - huh.
|
and for unvoiced it 's something that 's more flat.
|
Uh - huh. Right.
| false |
QMSum_120
|
Uh - huh.
|
and for unvoiced it 's something that 's more flat.
|
Uh - huh. Right.
|
And the way to do this is that well , we have the we have the FFT because it 's computed in in the in the system , and we have the mel filter banks ,
| false |
QMSum_120
|
and for unvoiced it 's something that 's more flat.
|
Uh - huh. Right.
|
And the way to do this is that well , we have the we have the FFT because it 's computed in in the in the system , and we have the mel filter banks ,
|
Mm - hmm. Mm - hmm.
| false |
QMSum_120
|
Uh - huh. Right.
|
And the way to do this is that well , we have the we have the FFT because it 's computed in in the in the system , and we have the mel filter banks ,
|
Mm - hmm. Mm - hmm.
|
and so if we if we , like , remove the mel filter bank from the FFT , we have something that 's close to the excitation signal.
| false |
QMSum_120
|
And the way to do this is that well , we have the we have the FFT because it 's computed in in the in the system , and we have the mel filter banks ,
|
Mm - hmm. Mm - hmm.
|
and so if we if we , like , remove the mel filter bank from the FFT , we have something that 's close to the excitation signal.
|
Oh.
| false |
QMSum_120
|
Mm - hmm. Mm - hmm.
|
and so if we if we , like , remove the mel filter bank from the FFT , we have something that 's close to the excitation signal.
|
Oh.
|
It 's something that 's like a a a train of p a pulse train for voiced sound
| false |
QMSum_120
|
and so if we if we , like , remove the mel filter bank from the FFT , we have something that 's close to the excitation signal.
|
Oh.
|
It 's something that 's like a a a train of p a pulse train for voiced sound
|
OK.
| false |
QMSum_120
|
Oh.
|
It 's something that 's like a a a train of p a pulse train for voiced sound
|
OK.
|
Yeah.
| false |
QMSum_120
|
It 's something that 's like a a a train of p a pulse train for voiced sound
|
OK.
|
Yeah.
|
Oh ! OK. Yeah.
| false |
QMSum_120
|
OK.
|
Yeah.
|
Oh ! OK. Yeah.
|
and that 's that should be flat for
| false |
QMSum_120
|
Yeah.
|
Oh ! OK. Yeah.
|
and that 's that should be flat for
|
Yeah.
| false |
QMSum_120
|
Oh ! OK. Yeah.
|
and that 's that should be flat for
|
Yeah.
|
I see. So do you have a picture that sh ?
| false |
QMSum_120
|
and that 's that should be flat for
|
Yeah.
|
I see. So do you have a picture that sh ?
|
So - It 's Y
| false |
QMSum_120
|
Yeah.
|
I see. So do you have a picture that sh ?
|
So - It 's Y
|
Is this for a voiced segment ,
| false |
QMSum_120
|
I see. So do you have a picture that sh ?
|
So - It 's Y
|
Is this for a voiced segment ,
|
yeah.
| false |
QMSum_120
|
So - It 's Y
|
Is this for a voiced segment ,
|
yeah.
|
this picture ? What does it look like for unvoiced ?
| false |
QMSum_120
|
Is this for a voiced segment ,
|
yeah.
|
this picture ? What does it look like for unvoiced ?
|
Yeah.
| false |
QMSum_120
|
yeah.
|
this picture ? What does it look like for unvoiced ?
|
Yeah.
|
You have several some unvoiced ?
| false |
QMSum_120
|
this picture ? What does it look like for unvoiced ?
|
Yeah.
|
You have several some unvoiced ?
|
The dif No. Unvoiced , I don't have
| false |
QMSum_120
|
Yeah.
|
You have several some unvoiced ?
|
The dif No. Unvoiced , I don't have
|
Oh.
| false |
QMSum_120
|
You have several some unvoiced ?
|
The dif No. Unvoiced , I don't have
|
Oh.
|
for unvoiced.
| false |
QMSum_120
|
The dif No. Unvoiced , I don't have
|
Oh.
|
for unvoiced.
|
Yeah. So , you know , all
| false |
QMSum_120
|
Oh.
|
for unvoiced.
|
Yeah. So , you know , all
|
I 'm sorry.
| false |
QMSum_120
|
for unvoiced.
|
Yeah. So , you know , all
|
I 'm sorry.
|
But Yeah.
| false |
QMSum_120
|
Yeah. So , you know , all
|
I 'm sorry.
|
But Yeah.
|
Yeah.
| false |
QMSum_120
|
I 'm sorry.
|
But Yeah.
|
Yeah.
|
Yeah. This is the between
| false |
QMSum_120
|
But Yeah.
|
Yeah.
|
Yeah. This is the between
|
This is another voiced example. Yeah.
| false |
QMSum_120
|
Yeah.
|
Yeah. This is the between
|
This is another voiced example. Yeah.
|
No. But it 's this ,
| false |
QMSum_120
|
Yeah. This is the between
|
This is another voiced example. Yeah.
|
No. But it 's this ,
|
Oh , yeah. This is
| false |
QMSum_120
|
This is another voiced example. Yeah.
|
No. But it 's this ,
|
Oh , yeah. This is
|
but between the frequency that we are considered for the excitation
| false |
QMSum_120
|
No. But it 's this ,
|
Oh , yeah. This is
|
but between the frequency that we are considered for the excitation
|
Right. Mm - hmm.
| false |
QMSum_120
|
Oh , yeah. This is
|
but between the frequency that we are considered for the excitation
|
Right. Mm - hmm.
|
for the difference and this is the difference.
| false |
QMSum_120
|
but between the frequency that we are considered for the excitation
|
Right. Mm - hmm.
|
for the difference and this is the difference.
|
Yeah.
| false |
QMSum_120
|
Right. Mm - hmm.
|
for the difference and this is the difference.
|
Yeah.
|
This is the difference. OK.
| false |
QMSum_120
|
for the difference and this is the difference.
|
Yeah.
|
This is the difference. OK.
|
So , of course , it 's around zero ,
| false |
QMSum_120
|
Yeah.
|
This is the difference. OK.
|
So , of course , it 's around zero ,
|
Yeah.
| false |
QMSum_120
|
This is the difference. OK.
|
So , of course , it 's around zero ,
|
Yeah.
|
Sure looks
| false |
QMSum_120
|
So , of course , it 's around zero ,
|
Yeah.
|
Sure looks
|
but
| false |
QMSum_120
|
Yeah.
|
Sure looks
|
but
|
Hmm.
| false |
QMSum_120
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.