Oulipo S+7 constraint, implemented in Python 3 & NLTK

oulipo.py

# an implementation of Oulipo S+7
# see https://en.wikipedia.org/wiki/Oulipo

from nltk import pos_tag, word_tokenize
from nltk.corpus import wordnet

text = "I am sitting in a room different from the one you are in now. I am recording the sound of my speaking voice and I am going to play it back into the room again and again until the resonant frequencies of the room reinforce themselves so that any semblance of my speech, with perhaps the exception of rhythm, is destroyed. What you will hear, then, are the natural resonant frequencies of the room articulated by speech. I regard this activity not so much as a demonstration of a physical fact, but more as a way to smooth out any irregularities my speech might have."

# download nltk tokeniser if missing
try:
    pos_tag(word_tokenize(text))
except LookupError:
    import nltk
    nltk.download( 'punkt' )
    nltk.download( 'averaged_perceptron_tagger' )

# tokenise & POS-tag quotation
tags = pos_tag(word_tokenize(text))

# check we've got the wordnet corpus, download if missing
try:
    wordnet.all_synsets()
except LookupError:
    import nltk
    nltk.download('wordnet')

# construct an alphabetised list of nouns from wordnet
# TODO: make this less stupid
nouns = []
for synset in list(wordnet.all_synsets('n')):
    name = synset.name().split(".")[0]
    if "_" not in name:
        nouns.append( name )
nouns.sort()

# step through tokenised quotation, replace nouns (ouilpo +7)
# https://en.wikipedia.org/wiki/Oulipo
words_out = []
for word, tag in tags:
    word_out = word
    # if word is a noun...
    if 'n' in tag.lower():
        if word in nouns:
            i = nouns.index( word )
            i = min( i+7, len(nouns) )
            word_out = nouns[ i ]
    words_out.append( word_out )

# print mangled quotation
print( " ".join( words_out)  )