this is my first post (on video.stackexchange.com).
I need some explanation regarding to re-encoding a single .ts chunk and then use
I'll explain my scenario.
I've a lot of
.ts chunks that all together create a smooth video. Tested.
To create a single video I use the classic linux
cat command, and everything works a treat.
These "lot of
.ts chunks" reside on a web-hosting where, obviously, it is not possible to install ffmpeg, but I can use
cat command (through PHP's
is to apply an audio/video fade-in on first chunk (and so far I've done it with
-vf "fade=t=in:st=0.1:d=1" and
-af "afade=t=in:st=0.1:d=1") on my local machine and then, after uploaded the "faded chunk" on remote web-hosting,
cat the faded chunk + remaing chunks.
My actual results
My concatenated-chunks-video has a frame-gap at the end of the "faded chunk" and the beginning of the next chunk. After this, plays smoothly.
ffmpeg -i /path/to/first_chunk.ts -vf "fade=t=in:st=0.1:d=1, format=yuv420p" -af "afade=t=in:st=0.1:d=1" -c:v libx264 -b:v 5099.998K -profile:v main -level:v 4.0 -c:a aac -b:a 279000 -strict experimental -y /path/to/first_chunk_fade_in.ts cat /path/to/first_chunk_fade_in.ts /path/to/2nd_chunk.ts /path/to/3rd_chunk.ts /path/to/4th_chunk.ts /path/to/5th_chunk.ts > /path/to/merged.ts
-b:a values were recovered and / or calculated from the original chunk using
Any ideas? Or do I have to re-encode the whole file?