How to Split Large Audio Files for Salad Transcription API

Last Updated: February 28, 2025

Introduction

When using Salad Transcription API, audio files have a maximum length limit of 2.5 hours each. If you have audio files longer than this to transcribe, you will need to split these into shorter segments first.

---

Prerequisites

Before you begin, make sure you have:

1. Python Installed: Ensure you have Python 3.8 or higher.
2. Libraries Installed: Use the following command to install the required libraries:
   pip install pydub
   
3. FFmpeg Installed: FFmpeg is needed by pydub for processing audio:
   • Linux: sudo apt install ffmpeg
   • MacOS: brew install ffmpeg
   • Windows: Download FFmpeg and add it to your system PATH.

---

Last Updated: February 28, 2025

Splitting the audio files

1. Create the script:

Create a Python script named split_audio.py to split the audio file:

from pydub import AudioSegment
import os

def split_audio(file_path, output_dir, segment_length=int(2.5*60*60*1000)): ## Adjust this value if you want even shorter segments.
  audio = AudioSegment.from_file(file_path)
  total_length = len(audio)
  os.makedirs(output_dir, exist_ok=True)

  for i in range(0, total_length, segment_length):
    segment = audio[i:i+segment_length]
    file_constant = os.path.splitext(os.path.basename(file_path))[0]
    segment.export(os.path.join(output_dir, f"{file_constant}_segment_{i//segment_length + 1}.

if __name__ == "__main__":
  input_file = "path/to/file.mp3" ## Set this to the location of the file you want to split. MP3 and MP4 files are compatible, but will always convert to MP3.
  output_directory = "output/directory" ## Set this to the output directory you want to use.
  split_audio(input_file, output_directory)

This script will split your large audio file into smaller segments of 2.5 hours maximum each and save them in the specified output directory. You can input either audio or video files, and they will convert to MP3 when splitting. Make sure to set the input, and output, directories for your files.

2. Run the script:

python split_audio.py

After running the script, your output directory should now contain your original audio clip split into compatible segments to transcribe with Salad Transcription API. It should look something like this:

You can now use these audio files to transcribe with using Salad Transcription API.