LuluW8071/ASR-with-Speech-Sentiment-and-Text-Summarizer

Custom Script using FFMpeg and train-test split

Closed this issue · 1 comments

LuluW8071 commented

A custom script to convert lossy MP3 data to lossless WAV format and splitting them into train and test via JSON format

[
    {   
        "key": "/path/to/audio/speech.wav", 
        "text": "This is a sentence of converted speech audio."
    },
    ...
]
  • Works on linux
  • Works on Windows
LuluW8071 commented
  • Used SOX instead of FFMpeg to speedup the conversion process by multithreading
  • Conversion MP3 -> FLAC