Unfortunately, the quality of the transcoding process for medium quality videos (for mobile devices) is poor.
This tool aims to increase the quality lanching a optimized transcoding process base in the original video. The output can be H.264 HighProfile (best compression/quality rate) or the new H.265/HEVC codec.
IMPORTANT! Modify the configuration file video_quality_enhancer.conf
adding the necessary configuration. Mainly the name of the users, the address of the common folder (if it exists), and the path where the photos and videos of each of the users are stored. You can do it in the 4th installation step.
InputFolders | Explanation | Default value | Example |
---|---|---|---|
USERS | Users configured in DSM | user1, user2, user3 | jhon, marc, tonny |
VIDEO_PATH | Common folder for photos & videos | /volume1/photo/ | |
USER_VIDEO_PATH | Generic Path where private users fotos are stored | /volume1/homes/user/Photos/ |
Using the File Station APP in your Synology, create a directory to host the code (Example: /volume1/code/synology-transcoding/)
Download the source code from the last release version. https://github.com/cibrandocampo/synology-transcoding/releases
Unzip the folder
Configure the video_quality_enhancer.conf
adding the necessary configuration. More info at Configuration/Customization block.
Copy the uncompressed photo-enhancer
folder to the NAS folder recently created.
Schedule the project execution to run once or multiple times a day. For this use the DSM task manager. Go to Control Panel
> Task Scheduler
, click Create
, and select Scheduled Task
General
video quality enhancer
Schedule (When you whant to execute the process)
Task Settings
User-defined script, add:
cd /volume1/code/synology-transcoding/photo-enhancer/ && python video_quality_enhancer.py
NOTE: Use the path configured in the step 1.
Synology manual: https://kb.synology.com/en-uk/DSM/help/DSM/AdminCenter/system_taskscheduler?version=7
If you encounter any problems or have any suggestions, feel free to contact me via (hello@cibran.es). You can also contribute to improving this project by submitting pull requests.
This project is licensed under the MIT License.