data = []
   for i in d['items']:
       start_time = i["start_time"]
       end_time = i["end_time"]
          for k in i["alternatives"]:
              content = k["content"]
              data.append([start_time,end_time,content])

0 list 3 [6.367, 6.597, 'Hi']
     1 list 3 [None, None, ',']
     2 list 3 [6.597, 6.727, 'good']
     3 list 3 [6.727, 6.987, 'afternoon']
     4 list 3 [None, None, '.']
     5 list 3 [6.987, 7.117, 'Thank']

0 list 2 [0.7500000000000004, 'christian_agent']
     1 list 2 [4.949999999999997, 'christian_agent']
     2 list 2 [12.659999999999922, 'christian_agent']
     3 list 2 [21.57000000000006, 'christian_agent']
     4 list 2 [27.810000000000297, 'christian_agent']
     5 list 2 [36.75000000000063, 'christian_agent']
     6 list 2 [47.55000000000104, 'christian_agent']

0 float 1 1.350000000000001
     1 float 1 10.529999999999967
     2 float 1 16.28999999999986
     3 float 1 25.620000000000214
     4 float 1 29.820000000000373
     5 float 1 37.44000000000066
     6 float 1 48.21000000000107.

import numpy as np
import pandas as pd
df= pd.DataFrame(data)
df=df.dropna(how='any')
for start1,end1 in zip(start,end):
    start_time = start1
    end_time = end1
for i, row in df.iterrows():
    start_word=row[0]
    end_word=row[1]
    content=row[2]
    if start_time >= start_word and end_time <= end_word:
       content1 = np.where[start_time >= start_word and end_time<=end_word]