Using pydub.silence

圖三 Silence Removal using pydub split_on_silence function

#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""
Created on Thu Jul  5 16:21:29 2018

@author: PeterTsai
"""

from pydub import AudioSegment
from pydub.silence import split_on_silence
#import numpy as np
import glob
import re

sliencePeriod = 800
source_dir = 'choices'
summary_length_txt = 'summary_time_length_pydub{}.txt'.format(sliencePeriod)


speechFileList = sorted(glob.glob(source_dir+'/*.wav'))

summary_length_fid = open(summary_length_txt,'w')


for speechFile in speechFileList:
    print('{} processing...'.format(speechFile))
    speech = AudioSegment.from_wav(speechFile)

    #pattern = '\./'+ source_dir + '/(.+)\.wav'
    m = re.match('{}/(.+)\.wav'.format(source_dir), speechFile)
    filename = m.group(1)

    chunks = split_on_silence(speech, 
        # must be silent for at least 2 seconds or 2000 ms
        min_silence_len=sliencePeriod,

        # consider it silent if quieter than -16 dBFS
        #Adjust this per requirement
        silence_thresh=speech.dBFS
    )

    time_length = [len(chunk)/1000 for chunk in chunks]

    #time_str = ', '.join(time_length)
    time_str = str(time_length).strip('[]')
    summary_length_fid.write('{}: {} ({} elements)\n'.format(filename,time_str,len(time_length)))

summary_length_fid.close()

Figure 1$\colon$ flowchart of choices audio segment time intervals

Figure 2$\colon$ result of segments' time interval