audio
audioduration (s)
0.22
26.4
sampling_rate
int64
16k
16k
transcript
stringlengths
1
378
16,000
yeah
16,000
<LAUGH>
16,000
you know to death that really did not do it
16,000
umhum
16,000
no i have a two through you know two through wedge i hit a two iron awful lot off now
16,000
umhum
16,000
right
16,000
uh practically a suspended sentence i mean it was
16,000
yeah
16,000
then i had seen something on television
16,000
oh yeah that is always a challenge so do you do my wife talks with other people
16,000
the justice system works is they bend over backwards trying to protect the guilty so many things in their back pardon yes i know
16,000
RIGHT
16,000
hm
16,000
yeah
16,000
you know uh
16,000
she uh uh always our house borders richland college i mean you walk out our back door and you are on the campus
16,000
you know the individuals to start thinking about what they can do to help each other out instead of counting on government to do everything
16,000
you probably do not follow that
16,000
hum
16,000
yeah
16,000
and you do not have to worry about uh fertilizing which causes a lot of thatch and all that kind of stuff so there is more and more people too are aware of that up here
16,000
super vga and yeah it is fun i do programming too my job is uh edp auditor and so i am into computers all the time
16,000
but
16,000
sorry we have kind of an open house and the sound really travels um <LAUGH>
16,000
well what have you seen recently
16,000
have not got so fortunate as to get into any of those companies yet
16,000
um we took a golf clinic there was about i do not know twelve fifteen people in it
16,000
it is uh it is really bad here uh for example the uh local high school uh they have already found two students with sawed off shotguns
16,000
right
16,000
lot of WORK
16,000
right that is when you see everybody is when somebody dies it is awful <LAUGH>
16,000
is even on tv you watch it for two minutes and think this is so ridiculously stupid <LAUGH>
16,000
and uh a um a nineteen eleven which is fairly scarce
16,000
oh yeah yeah yeah
16,000
oh how fun oh that is fun
16,000
right
16,000
are you uh and that my walking was a fast walk uh to get the heart rate up to an aerobic level
16,000
yeah right
16,000
down right ridiculous and i know driving through pittsburgh you know we got lost DOWN THERE ONE TIME and we we are trying to come home and um
16,000
yeah
16,000
uh near here and uh so but the houses around here there is five five houses on my street for sale right now
16,000
english comedies
16,000
umhum
16,000
wow uhhuh
16,000
i do not know i
16,000
i think a lot of the commentators on like the major networks like right it is kind of appropriate right now because of the election stuff going on but um it seems that um
16,000
okay uh mandatory service yeah i do not think it is a good idea
16,000
i have i have done several of them it is it is it is all right it is you know
16,000
yes
16,000
yes
16,000
yeah
16,000
oh my gosh
16,000
YEAH I KNOW know then WHEN one of the doors fall off then it is time to get a new car YOU KNOW
16,000
yeah sounds like you have a real good garden
16,000
well i guess we are supposed to talk about
16,000
all right
16,000
and
16,000
yeah well
16,000
oh really
16,000
uh and and find out every
16,000
well at least you will get very good at it right
16,000
and they did not get across the river there so they are the last and that was actually after the war was over but it was the last
16,000
i like the music but i have been unable to do that because i hurt my foot about five years ago i broke my heel
16,000
yeah
16,000
right
16,000
he said he would lived on a farm too long and that was too much like camping ANYWAY true camping
16,000
no <LAUGH> my mom has a cat though that does that
16,000
when i can roll that old beast of mine down there for a lot less than that
16,000
it it was fascinating it really was i really enjoyed it and i think everyone else did otherwise we would not keep doing it every morning
16,000
yeah
16,000
uhhuh
16,000
i do not and i have an older sister that loves running too and she runs all the time
16,000
<LAUGH>
16,000
uhhuh
16,000
a chicken
16,000
yeah uhhuh ME TOO i am not sure that i use them as often as i think that i will but i sure uh pay attention to them and cut them out
16,000
ouch
16,000
you know men and women alike they give them say three hundred bucks a month and they can buy their health insurance and their retirement and whatever but
16,000
so uh just a just a change of pace
16,000
umhum that is true
16,000
right
16,000
uhhuh
16,000
five pounds or
16,000
oh sure
16,000
oh
16,000
took that uh went to war with iraq i think we that was kind of a uh display of power
16,000
i do not i guess i could i have never been to a to a major golf tournament i uh watch them on tv a lot
16,000
right
16,000
right what kind of car is it
16,000
that that would be important for me that that there was not that it was optional that if if every person and i could pick a time i guess rather than worrying about whether it was one years or two years suppose it were to be uh eighteen months
16,000
umhum
16,000
yeah
16,000
uhhuh
16,000
i know i i had a college professor that went i went to school in in missouri that is where we are from and he went to dc on a for a a big conference he is a political science uh
16,000
fortunately
16,000
oh okay
16,000
yeah
16,000
it is really a sad situation
16,000
but there there definitely needs to be a BALANCE SOMEWHERE

A preprocessed version of Switchboard Corpus. The corpus audio has been upsampled to 16kHz, separated channels and the transcripts have been processed with special treats for paralinguistic events, particularly laughter and speech-laughs. This preprocessed dataset has been processed for ASR task. For the original dataset, please check out the original link: https://catalog.ldc.upenn.edu/LDC97S62

The dataset has been splitted into train, test and validation sets with 70/20/10 ratio, as following summary:

Train Dataset (70%): Dataset({
    features: ['audio', 'sampling_rate', 'transcript'],
    num_rows: 185402
})
Validation Dataset (10%): Dataset({
    features: ['audio', 'sampling_rate', 'transcript'],
    num_rows: 20601
})
Test Dataset (20%): Dataset({
    features: ['audio', 'sampling_rate', 'transcript'],
    num_rows: 51501
})

An example of the content is this dataset:


Regarding the total amount of laughter and speech-laugh existing in the dataset, here is the overview:

Train Dataset (swb_train): {'laughter': 16044, 'speechlaugh': 9586} 

Validation Dataset (swb_val): {'laughter': 1845, 'speechlaugh': 1133} 

Test Dataset (swb_test): {'laughter': 4335, 'speechlaugh': 2775} 
Downloads last month
16