| Season | Time |
|---|---|
| s1 | 09h 50m 07s |
| s2 | 14h 57m 31s |
| s3 | 13h 03m 22s |
| s4 | 13h 26m 00s |
| extras | 02h 02m 22s |
| TOTAL | 53h 19m 22s |
https://talesfromthecodemines.wordpress.com/2016/10/09/how-to-use-libvlc-on-windows/
Use gentle and google auto-subtitle somehow
Hash map to store word count total
SPELL CHECK
Load up clip with subtitles and have option to split clip equally via spaces
use trie
ffmpeg -i episode.mkv -vf scale=-1:720 -c:v libx264 -crf 0 -preset veryslow -c:a copy episode_720.mkv ffmpeg -i episode.mkv -filter:v scale=1280:-1 -c:a copy episode_720.mkv
Word
- id
- Begin timestamp
- End timestamp
PulseAudio hosted at 127.0.0.1 Display at port 0 export DISPLAY=:0 export PULSE_SERVER=tcp:127.0.0.1 export LIBGL_ALWAYS_INDIRECT=Yes
Labelling program
- get audio, video, and if present subtitle streams
- if subtitle streams, generate intervals with names
- Use Gentle to force align words -- Cut out audio -- Put in subtitle thing -- delete old interval, import new intervals -- adjust mistakes
- split evenly at spaces if gentle doesn't work
- Label words based on audio samples, frames are too imprecise!!
- Generate csv or whatever
- OVERALL STATS GETS SAVED SEPARATELY (can update by re-tallying all csvs in saved directory)
Compiling program
- concat all csvs together
- seperate into different letters (A B C, etc)
- order alphabetically
- go into each clip, trim, atrim, etc
- apply neat-looking overlays (calculate stats using global and letter csv)
- Pray nothing goes wrong when concat
SFML and sfeMovie!!!!!!
use wasdqe and ijkluo as two "joysticks" ctrl + g to perform gentle analysis