text_cloud.py

import os
import pyopenjtalk
import re

out_path = './text_cloud.txt'
output = []
j=0
in_path = '/mnt/vol21/i21_mkazuki/text/transcripts_utf8.txt'
for i in range(1,101):
    if i in [1,3,5,6,9,11,12,13,20,21,22,23,28,31,32,33,34,37,41,42,44,45,46,47,48,49,50,52,54,68,70,71,73,74,75,76,77,78,79,80,81,86,87,88,89,97,98,99,100]:
        continue
    with open(in_path) as f:
        lines = f.readlines()
        for line in lines:
            strs = line.split(':')
            if not (os.path.isfile('/mnt/nfs-mnj-archive-03/group/creative/i21_mkazuki/wav/jvs'+format(i, '03')+'/parallel100/'+strs[0]+'.wav')):
                continue
            strs[1] = pyopenjtalk.g2p(strs[1])
            strs[1] = strs[1].replace('pau',',')
            
    nonpara_path = '/mnt/nfs-mnj-archive-03/group/creative/i21_mkazuki/wav/jvs_ver1/jvs'+format(i, '03')+'/nonpara30/transcripts_utf8.txt'
    with open(nonpara_path) as f:
        lines = f.readlines()
        for line in lines:
            strs = line.split(':')
            if not (os.path.isfile('/mnt/nfs-mnj-archive-03/group/creative/i21_mkazuki/wav/jvs'+format(i, '03')+'/nonpara30/'+strs[0]+'.wav')):
                continue
            strs[1] = pyopenjtalk.g2p(strs[1])
            strs[1] = strs[1].replace('pau',',')
            
            if j==0:
                output.append('/home/mkazuki/eval_wav/'+strs[0]+'.wav|'+strs[1]+'|'+str(j)+'\n')
            
    whisper_path = '/mnt/nfs-mnj-archive-03/group/creative/i21_mkazuki/wav/jvs_ver1/jvs'+format(i, '03')+'/whisper10/transcripts_utf8.txt'
    with open(whisper_path) as f:
        lines = f.readlines()
        for line in lines:
            strs = line.split(':')
            if not (os.path.isfile('/mnt/nfs-mnj-archive-03/group/creative/i21_mkazuki/wav/jvs'+format(i, '03')+'/whisper10/'+strs[0]+'.wav')):
                continue
            strs[1] = pyopenjtalk.g2p(strs[1])
            strs[1] = strs[1].replace('pau',',')
            
            if j==0:
                output.append('/home/mkazuki/eval_wav/whisper_'+strs[0]+'.wav|'+strs[1]+'|'+str(j)+'\n')
    j+=1
    
k=1

with open(in_path) as f:
    lines = f.readlines()
    for line in lines:
        strs = line.split(':')
        strs[1] = pyopenjtalk.g2p(strs[1])
        strs[1] = strs[1].replace('pau',',')
        
        if k>95:
            output.append('/home/mkazuki/eval_wav/tsuchiya_normal_'+format(k, '03')+'.wav|'+strs[1]+'|'+str(j)+'\n')
        k+=1
        
k=1        
with open(in_path) as f:
    lines = f.readlines()
    for line in lines:
        strs = line.split(':')
        strs[1] = pyopenjtalk.g2p(strs[1])
        strs[1] = strs[1].replace('pau',',')
        
        if k>95:
            output.append('/home/mkazuki/eval_wav/tsuchiya_happy_'+format(k, '03')+'.wav|'+strs[1]+'|'+str(j)+'|2'+'\n')

        k+=1
        
k=1        
with open(in_path) as f:
    lines = f.readlines()
    for line in lines:
        strs = line.split(':')
        strs[1] = pyopenjtalk.g2p(strs[1])
        strs[1] = strs[1].replace('pau',',')
        
        if k>95:
            output.append('/home/mkazuki/eval_wav/tsuchiya_angry_'+format(k, '03')+'.wav|'+strs[1]+'|'+str(j)+'|3'+'\n')
        k+=1    
j+=1
k=1
with open(in_path) as f:
    lines = f.readlines()
    for line in lines:
        strs = line.split(':')
        strs[1] = pyopenjtalk.g2p(strs[1])
        strs[1] = strs[1].replace('pau',',')
        
        if k>95:
            output.append('/home/mkazuki/eval_wav/uemura_normal_'+format(k, '03')+'.wav|'+strs[1]+'|'+str(j)+'\n')
        k+=1
        
k=1        
with open(in_path) as f:
    lines = f.readlines()
    for line in lines:
        strs = line.split(':')
        strs[1] = pyopenjtalk.g2p(strs[1])
        strs[1] = strs[1].replace('pau',',')
        
        if k>95:
            output.append('/home/mkazuki/eval_wav/uemura_happy_'+format(k, '03')+'.wav|'+strs[1]+'|'+str(j)+'|2'+'\n')

        k+=1
        
k=1        
with open(in_path) as f:
    lines = f.readlines()
    for line in lines:
        strs = line.split(':')
        strs[1] = pyopenjtalk.g2p(strs[1])
        strs[1] = strs[1].replace('pau',',')
        
        if k>95:
            output.append('/home/mkazuki/eval_wav/uemura_angry_'+format(k, '03')+'.wav|'+strs[1]+'|'+str(j)+'|3'+'\n')
        k+=1    

j+=1
k=1
with open(in_path) as f:
    lines = f.readlines()
    for line in lines:
        strs = line.split(':')
        strs[1] = pyopenjtalk.g2p(strs[1])
        strs[1] = strs[1].replace('pau',',')
        
        if k>95:
            output.append('/home/mkazuki/eval_wav/fujitou_normal_'+format(k, '03')+'.wav|'+strs[1]+'|'+str(j)+'\n')
        k+=1
        
k=1        
with open(in_path) as f:
    lines = f.readlines()
    for line in lines:
        strs = line.split(':')
        strs[1] = pyopenjtalk.g2p(strs[1])
        strs[1] = strs[1].replace('pau',',')
        
        if k>95:
            output.append('/home/mkazuki/eval_wav/fujitou_happy_'+format(k, '03')+'.wav|'+strs[1]+'|'+str(j)+'|2'+'\n')
        k+=1
        
k=1        
with open(in_path) as f:
    lines = f.readlines()
    for line in lines:
        strs = line.split(':')
        strs[1] = pyopenjtalk.g2p(strs[1])
        strs[1] = strs[1].replace('pau',',')
        
        if k>95:
            output.append('/home/mkazuki/eval_wav/fujitou_angry_'+format(k, '03')+'.wav|'+strs[1]+'|'+str(j)+'|3'+'\n')
        k+=1    
        
        
with open(out_path, 'w') as f:
    f.writelines(output)