video-transcode

Simplified commercial cutting and transcoding for Plex DVR.

video_transcode removes commercials and transcodes Plex recorded TV shows to smaller file. A 60 minute DVR show starting at ~5GB mpeg2 .ts file is reduced to a ~40-42 minute ~800MB h.265 .mkv file with little quality loss.

The service also supports Nvidia GPU transcoding, controls how intensive transcoding is on a machine and what time of day trnascoding can happen. e.g. only transcode overnight, limit 1 transcode at a time, etc.

• Quickstart
• How do I use this?
• Docker Setup
• Configuration Options
• Manual installation

Quickstart

1. Install video_transcode with pip install video_transcode

2. Create docker-compose.yaml

   version: '3'
   services:
       redis:
           restart: unless-stopped
           image: redis:alpine
           ports:
               - "6124:6379"
       video:
           image: kirb5/video-transcode
           restart: unless-stopped
           depends_on:
               - "redis"
           volumes: 
               # REQUIRED! Folder containing Plex DVR recordings on host must be maped to /home/plex inside container
               - ./plex:/home/plex
   
   

3. Start docker-compose up

4. Add a video file to the transcoding queue with video-transcode "MacGyver - S04E01 - Fire + Ashes + Legacy = Phoenix.ts"

Usage

Basic usage

video-transcode input_file.ts

Plex postprocessing setup

Configuring Plex to use video-transcode for post processing requires:

1. Symlink video-transcode executable to DVR post processing script folder. The exact path to this folder is visible within Plex DVR settings under Postprocessing Script (see red highlight below).

$ ln -s $( which video-transcode ) /path/to/plex/Library/Application/Support/Plex\ Media\ Server/Scripts/video-transcode

2. Add absolute path of symlink to Postprocessing Script section in DVR settings (see orange highlight below).

Docker

Docker image is prebuilt with Nvidia HEVC and AVC encoding/decoding support for GPU accellerated transcodes. The image requires access to a Redis server. Specify the server's URL in CELERY_BROKER and CELERY_RESULT_BACKEND keys in video_transcode/config/config.yaml

Nvidia GPU Transcoding (NVENC)

Prerequisites

Docker 19.03+ and (nvidia-container-toolkit)[https://github.com/NVIDIA/nvidia-docker] are required for GPU transcoding. ffmpeg built with cuvid and nvenc support using nv-codec-headers 9.1.23.1, CUDA 10.2 and Nvidia driver 440.82.

Setup

As of 7/2/20 docker-compose does not support --gpus flag in Docker 19.03. Manually setting up the service is the only option for now.

1. Install video_transcode with pip
2. Create network in docker
3. Create redis container
4. Create video_transcode container

Configuration

video_transcode

celery

Manual Installation

1. Install ffmpeg
2. Install redis
3. Install comchap
4. Install comskip
5. Install video-transcode
6. make
7. start service
8. Add video-transcode script path to Plex post processing.

Architecture

video_transcode stack includes:

• ffmpeg
• Compskip/comchap
• celery
• redis

To Do

• Configurable UID/GID
• Pass additional ffmpeg options
• Port comcut to Python