Skip to content

Quick Start

Get started with S2ST Translator in just a few minutes.

Terminal window
# Start both the worker and web server
./start.sh

This will:

  1. Activate the virtual environment
  2. Update dependencies
  3. Build the frontend
  4. Start the background worker
  5. Start the web server on port 8000

Open http://localhost:8000 in your browser.

  1. Click “New Translation” in the dashboard
  2. Choose your input method:
    • Upload Files - Drag and drop audio/video files
    • Local Path - Enter a path on the server
    • YouTube URL - Paste a YouTube video or playlist URL
  3. Select the target language
  4. Click Submit
  • View real-time progress in the dashboard
  • Check job details by clicking the view icon
  • Download completed translations
Terminal window
./run.sh translate --input audio.wav --target-lang deu
Terminal window
./run.sh job submit --input audio.wav --target-lang fra
Terminal window
./run.sh job list
Terminal window
./run.sh info

=== “Web Dashboard”

  1. Click “New Translation”
  2. Select “YouTube URL” tab
  3. Paste the video URL
  4. Select target language
  5. Submit and wait for completion

=== “CLI”

The CLI currently requires downloading first:

Terminal window
# Download with yt-dlp
yt-dlp -x --audio-format wav "https://youtube.com/watch?v=..."
# Translate
./run.sh translate --input video.wav --target-lang jpn

=== “Web Dashboard”

  1. Click “New Translation”
  2. Drag multiple files into the upload zone
  3. Select target language
  4. Submit as batch

=== “CLI”

Terminal window
# Process all files in a directory
for f in input/*.wav; do
./run.sh job submit --input "$f" --target-lang deu
done

When you upload a video file (MP4, MKV, etc.), S2ST Translator will:

  1. Extract the audio track
  2. Translate the audio
  3. Re-mux the translated audio back into the video

The output includes both:

  • filename_translated.mp4 - Video with translated audio
  • filename_translated.wav - Standalone translated audio