TempoCut

Broadcast-style video & audio time compression — recreate the classic “time tailoring” used by TBS, TNT, TruTV, Cartoon Network, and Adult Swim.
Not affiliated with Warner Bros. Discovery or Prime Image. Educational use only.

---

📖 Overview

TempoCut is a Python toolkit that mimics professional broadcast time compression systems (e.g., Prime Image Time Tailor). It shortens shows to fit time slots while keeping tight A/V sync.

The audio engine is designed to sound very close to the Turner “skippy” style with minimal artifacts, while the video is retimed to match (59.94p output with subtle smears). Subtitle alignment is handled automatically via DTW warp maps.

---

✨ Features

• Broadcast-accurate time compression for audio + video
• Multiple audio “skippy” modes with marker export for Premiere Pro
• 29.97p or 59.94p output with micro-smear blending to reduce judder
• Saves DTW warp map for subtitle retiming
• Windows batch pipeline for one-click runs

---

🛠 Requirements

Python: 3.10+

Install tools & libs

pip install -r requirements.txt

Also required

• FFmpeg in your PATH (for I/O and muxing)
• (Optional) ffsubsync CLI in PATH if you want the batch script’s subtitle fallback

---

🚀 Usage

TempoCut has three main stages: audio compression, video retiming, and subtitle alignment. You can run them individually or together with the provided batch script.

1. Audio Compression

Choose stereo or surround based on your source.

Stereo

python audio_skippy_STEREO.py -i "input.wav" -o "output.wav" --target-ratio 1.02

Surround (5.1 WAV)

python audio_skippy_SURROUND.py -i "input.wav" -o "output.wav" --target-ratio 1.02

👉 This step creates both the compressed audio file and a *_markers.txt file listing “skippy” points, which you can import into Premiere Pro.

---

2. Video Retime

Now retime the video to match the skippy audio.

python time_compressor_SAFE.py -i "input.mp4" -s "output.wav" -o "out_59p.mp4"

• Output is 29.97p or 59.94p video with micro-smear blending (to hide jumps).
• A warp map file map_t_skip_to_t_orig.npy is also created — you’ll need it if you want subtitles.

---

3. Subtitle Alignment

If you have subtitles, retime them using the warp map:

python retime_subs.py map_t_skip_to_t_orig.npy input.srt output.srt

This adjusts every subtitle cue to stay in sync with the new compressed video.

---

4. One-Click Workflow (Windows)

Edit the paths inside:

time_compressor_pipeline.bat

Then run it. It will:

1. Retime video
2. Mux audio
3. Retime subs (or fall back to ffsubsync)
4. Clean up temp files

---

🎚️ Audio Compression Modes

Basic usage

python audio_skippy.py -i "input.wav" -o "output_timecompressed.wav" --target-ratio 1.0198

(Change --target-ratio to your desired total compression.)

Advanced usage (classic “skippy” cadence)

python audio_skippy.py ^
 -i "input.wav" ^
 -o "output_compress.wav" ^
 --target-ratio **** ^
 --frame-ms 15 ^
 --max-chop-ms 35 ^
 --cadence-ms 250 ^
 --crossfade-ms 6 ^
 --energy-quantile 0.5

Lighter compression (smoother)

python audio_skippy.py ^
 -i "input.wav" ^
 -o "output_light.wav" ^
 --target-ratio **** ^
 --frame-ms 20 ^
 --max-chop-ms 25 ^
 --cadence-ms 300 ^
 --crossfade-ms 8 ^
 --energy-quantile 0.4

Heavier compression (more TBS-like “skips”)

python audio_skippy.py ^
 -i "input.wav" ^
 -o "output_heavy.wav" ^
 --target-ratio **** ^
 --frame-ms 10 ^
 --max-chop-ms 45 ^
 --cadence-ms 180 ^
 --crossfade-ms 4 ^
 --energy-quantile 0.6

Tips

• --target-ratio ≈ total shortening (e.g., 1.02 ≈ 2% shorter).
• Smaller --frame-ms/--cadence-ms = tighter sync, more obvious “skips”.
• Larger values = smoother, lighter compression.
• Keep ratios under 1.05 for natural sound.
• For 5.1, use audio_skippy_SURROUND.py.

---

⚠️ Known Issues & Workarounds

• Brief freeze at start if the first audio/video frames are silent/black.
  Workaround: Trim a tiny leading sliver (100–300 ms) before processing.
• Occasional mid-video frame pauses if off-by-one sync drift appears.
  Workaround: Try a larger --frame-ms or gentler --target-ratio on audio.
• Library compatibility
  Pinned: moviepy==1.0.5, numpy<2.0.

---

📂 Scripts in this repo

• audio_skippy_STEREO.py – stereo micro-skip engine + Premiere markers
• audio_skippy_SURROUND.py – multichannel/5.1 micro-skip engine + markers
• time_compressor_SAFE.py – DTW-based video retime to skippy audio (59.94p + smears) and saves warp map
• retime_subs.py – retimes SRTs via the saved warp map
• time_compressor_pipeline.bat – Windows pipeline for the whole flow

---

📜 License

MIT License — see LICENSE for details.