Sentence Segmentation

(in main.py)

43      song = AudioSegment.from_wav(speech)
44  
45      #split track where silence is 1 seconds or more and get chunks
46  
47      chunks = split_on_silence(song, 
48          # must be silent for at least 1 seconds or 1000 ms
49          min_silence_len=1000,
50  
51          # consider it silent if quieter than -16 dBFS
52          #Adjust this per requirement
53          silence_thresh=song.dBFS
54      )

split_on_silence是來自pydub.silence module 的function

Figure 3: Choice Audio Type [Option1]+[Option2]+[Option3]+[Option4]

Figure 3 Choice 音頻檔的波形示意圖，每個選項中間會有個比較長slience時間，這些temporal slience可以拿來當作Sentence Segmentation的evidence。值得注意的是這些temporal slience一樣也可以拿來當作Folder A裡面story audio file裡面當作斷局的evidence。