diff --git a/README.md b/README.md index 678b218..07433b2 100644 --- a/README.md +++ b/README.md @@ -88,21 +88,19 @@ With your own parameters: * [PySceneDetect](https://pyscenedetect.readthedocs.io/en/latest/) used for spliting video by scenes and running multiple encoders. -* Both Video and Image encoding* (Single frame to .ivf, can be viewed by videoplayer) +* Both Video and Image encoding* (Single frame to .ivf, can be viewed by videoplayer). -* Resuming encoding without loss of encoded progress +* Resuming encoding without loss of encoded progress. -* Simple and clean console look +* Simple and clean console look. -* Automatic determination of how many workers the host can handle +* Automatic determination of how many workers the host can handle. -* Building encoding queue with bigger files first, minimizing waiting for last scene to encode +* Building encoding queue with bigger files first, minimizing waiting for last scene to encode. -* Both video and audio encoding option with FFmpeg +* Both video and audio encoding option with FFmpeg. -* Logging of progress of all encoders - -And many more to go.. +* Logging of progress of all encoders. ## Install on Windows