

In [1]:

    
%pylab inline
# only display first 30 seconds
from pyannote.core.notebook import set_notebook_crop
from pyannote.core import Segment
set_notebook_crop(Segment(0, 30))









    



Populating the interactive namespace from numpy and matplotlib

The Big Bang Theory TVD Plugin

Install

The following command will install The Big Bang Theory TVD plugin (and TVD if it is missing)

pip install TVDTheBigBangTheory

Download all resources

The following command will download all resources for The Big Bang Theory into /tmp/ directory.

python -m tvd.create metadata /tmp/ TheBigBangTheory

Available resources

Initialize TheBigBangTheory TVD plugin



In [2]:

    
from tvd import TheBigBangTheory
theBigBangTheory = TheBigBangTheory('/tmp')









    



IN CASE YOU USE 'speaker' RESOURCES, PLEASE CONSIDER CITING:
@inproceedings{Tapaswi2012
    title = {{``Knock! Knock! Who is it?'' Probabilistic Person Identification in TV Series}},
    author = {Makarand Tapaswi and Martin B\"{a}uml and Rainer Stiefelhagen},
    booktitle = {IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
    year = {2012},
    month = {June},
}


IN CASE YOU USE 'outline' RESOURCES, PLEASE CONSIDER CITING:
@misc{the-big-bang-theory.com,
    title = {{The Big Bang Theory Wiki}},
    howpublished = \url{http://wiki.the-big-bang-theory.com/}
}


IN CASE YOU USE 'transcript' RESOURCES, PLEASE CONSIDER CITING:
@misc{bigbangtrans,
    title = {{big bang theory transcripts}},
    howpublished = \url{http://bigbangtrans.wordpress.com/}
}


IN CASE YOU USE 'transcript_www' RESOURCES, PLEASE CONSIDER CITING:
@misc{bigbangtrans,
    title = {{big bang theory transcripts}},
    howpublished = \url{http://bigbangtrans.wordpress.com/}
}


IN CASE YOU USE 'outline_www' RESOURCES, PLEASE CONSIDER CITING:
@misc{the-big-bang-theory.com,
    title = {{The Big Bang Theory Wiki}},
    howpublished = \url{http://wiki.the-big-bang-theory.com/}
}

Get first episode



In [3]:

    
episode = theBigBangTheory.episodes[0]
episode









    Out[3]:





Episode(series='TheBigBangTheory', season=1, episode=1)

Outlines

Source: http://wiki.the-big-bang-theory.com
Provides

scene summary
scene location



In [9]:

    
outline = theBigBangTheory.get_resource('outline', episode)
outline









    Out[9]:

Transcripts

Source: http://foreverdreaming.org/
Provides:

speaker label
speech content

Does not provide timestamps.



In [6]:

    
transcript = theBigBangTheory.get_resource('transcript', episode)
transcript









    Out[6]:

Force-aligned transcripts

Source: LIMSI
Provides:

word-level timestamps (start & end time)



In [7]:

    
transcript_aligned = theBigBangTheory.get_resource('transcript_aligned', episode)
transcript_aligned.crop(0, 30)









    Out[7]:

Speaker

Source: Makarand Tapaswi
Provides:

speaker label with manual timestamps (five main characters + other)



In [10]:

    
speaker = theBigBangTheory.get_resource('speaker', episode)
speaker









    Out[10]: