IDEA: 1. Skip intro & outro - X & Y seconds 2. Skip common phrases
Opened this issue · 4 comments
Describe the solution you'd like
A clear and concise description of what you want to happen.
Repetitive video intros on videos are annoying, most videos can have the first 30 seconds skipped with no ill effect. Perhaps even add an option to skip X seconds of the end as well to make playlists less annoying.
Also save your ears since many video creators have loud music during intro/outro sequences. Videos have been padded since the time youtube started favoring 10+ minute videos.
Thank you, there are several extensions allowing to skip seconds or percentage.
(We tend to work on things first, that don't seem not to exist yet.)
- Common phrases on youtube and in youtube categories can be identified & skipped by transcript
- Common phrases, slogans, etc, per channel can also be indentifies & skipped
https://docs.google.com/spreadsheets/d/1awdy0UvhaAz3ll1eexRSxNTmtpGbdBUYjgeiBDkoJI8/edit#gid=1433029918
- Common phrases, slogans, etc, per channel can also be indentifies & skipped
- skipping intro/outro
-
And we could integrate data like! SB / [NB]
(https://github.com/andrewzlee/NeuralBlock) etc. (Besides Sponsorblock's definition of intro might be animation only/shorter than yours #927 ) -
To identify intro&outro channel-wide habits, we could also check if there are typical durations of intro and outro length per channel / category in the top 5000 channels. Based on sponsorblock data or manually testing new and old videos from the same channel.
-
skipping intro: Bold automatic/experimental approach, for getting to the gist: (not what you said :D)
A rule of thumb is that that the first 30% are 'intro' contentually https://knowyourmeme.com/memes/the-wadsworth-constant . so this could be refined as follows:
Video will can start at high speed 3x and volume spikes capped (scalped).
Then gradually slow down to to reach normal speed at 30%.
This formular can be tuned and adjusted by current category or channel,
(or be dynamic per video, considering keywords in speech/captions, linguistic analysis of captions.)
-
"markersList": {
"markerType": "MARKER_TYPE_HEATMAP",
"markers": [
{
"startMillis": "0",
"durationMillis": "1234",
"intensityScoreNormalized": 0.1
},
"markersList": {
"markerType": "MARKER_TYPE_HEATMAP",
"markers": [
{
"startMillis": "0",
"durationMillis": "1234",
"intensityScoreNormalized": 0.1
},