ffmpeg creating image slideshow in audio concat command - audio

Question

I have a working ffmpeg which when ran in windows 10 command prompt will combine 3 audio files with 1 image file into a video.
The single image file is used as just a static image for the entire video's duration. (16:50 or 1010 seconds)
song1.mp3 00:00:00-00:06:23
song2.mp3 00:06:23-00:12:04
song3.wav 00:12:04-00:16:50
I am trying to change this so that instead of a single static image, I have a slideshow of 4 different images play during the duration of the video.
I was thinking of doing this by taking the total length in seconds (1010) divided by 4 would mean 252.5 seconds per image, so a timeline like:
img1: 0 - 252.5
img2: 252.5 - 505
img3: 505 - 757.5
img4: 757.5 - 252.5
How can I change my command so that the output video is a slideshow of images with the timestamps above?
Here is my working ffmpeg command with explanation from myself which hopefully explains some of it :
ffmpeg -loop 1 -framerate 2 -i "C:\Users\...local_filepath...\ffmpeg-commands\images\img1.png" -i "C:\Users\...local_filepath...\ffmpeg-commands\audio files\song1.mp3" -i "C:\Users\...local_filepath...\ffmpeg-commands\audio files\song2.mp3" -i "C:\Users\...local_filepath...\ffmpeg-commands\audio files\song3.wav" -c:a pcm_s32le -filter_complex concat=n=3:v=0:a=1 -vcodec libx264 -bufsize 3M -filter:v "scale=w=640:h=638,pad=ceil(iw/2)*2:ceil(ih/2)*2" -crf 18 -pix_fmt yuv420p -shortest -tune stillimage -t 1010 "C:\Users\...local_filepath...\ffmpeg-commands\videos\outputvideo.mkv"
ffmpeg
//loop single image
-loop 1
//set video framerate
-framerate 2
//take image and audio files
-i "C:\Users\...local_filepath...\ffmpeg-commands\images\img1.png"
-i "C:\Users\...local_filepath...\ffmpeg-commands\audio files\song1.mp3"
-i "C:\Users\...local_filepath...\ffmpeg-commands\audio files\song2.mp3"
-i "C:\Users\...local_filepath...\ffmpeg-commands\audio files\song3.wav"
//set audio codec
-c:a pcm_s32le
//organize the audio files (?)
-filter_complex concat=n=3:v=0:a=1
//set video codec
-vcodec libx264
//limiting the output bitrate
-bufsize 3M
//scale single image to 640x638 resolution
-filter:v "scale=w=640:h=638,pad=ceil(iw/2)*2:ceil(ih/2)*2"
//range of the CRF scale is 0–51, where 0 is lossless (for 8 bit only, for 10 bit use -qp 0), 23 is the default, and 51 is worst quality
-crf 18
//set pixel format
-pix_fmt yuv420p
//try to ensure video is not longer then concatenated audio files
-shortest
//use still image
-tune stillimage
//specify output video time to ensure it is exactly the same length as concattenated audio
-t 1010
//specify output filepath
"C:\Users\...local_filepath...\ffmpeg-commands\videos\outputvideo.mkv"
And here is the command raw how you would run it.
ffmpeg -loop 1 -framerate 2 -i "C:\Users\marti\Documents\projects\jan2023-rendertune\ffmpeg-commands\images\img1.png" -i "C:\Users\marti\Documents\projects\jan2023-rendertune\ffmpeg-commands\audio files\song1.mp3" -i "C:\Users\marti\Documents\projects\jan2023-rendertune\ffmpeg-commands\audio files\song2.mp3" -i "C:\Users\marti\Documents\projects\jan2023-rendertune\ffmpeg-commands\audio files\song3.wav" -c:a pcm_s32le -filter_complex concat=n=3:v=0:a=1 -vcodec libx264 -bufsize 3M -filter:v "scale=w=640:h=638,pad=ceil(iw/2)*2:ceil(ih/2)*2" -crf 18 -pix_fmt yuv420p -shortest -tune stillimage -t 1010 "C:\Users\marti\Documents\projects\jan2023-rendertune\ffmpeg-commands\videos\concatVideo-250771.mkv"
And here is my ffmpeg version info:
ffmpeg version 4.3 Copyright (c) 2000-2020 the FFmpeg developers
built with gcc 9.3.1 (GCC) 20200621
configuration: --enable-gpl --enable-version3 --enable-sdl2 --enable-fontconfig --enable-gnutls --enable-iconv --enable-libass --enable-libdav1d --enable-libbluray --enable-libfreetype --enable-libmp3lame --enable-libopencore-amrnb --enable-libopencore-amrwb --enable-libopenjpeg --enable-libopus --enable-libshine --enable-libsnappy --enable-libsoxr --enable-libsrt --enable-libtheora --enable-libtwolame --enable-libvpx --enable-libwavpack --enable-libwebp --enable-libx264 --enable-libx265 --enable-libxml2 --enable-libzimg --enable-lzma --enable-zlib --enable-gmp --enable-libvidstab --enable-libvmaf --enable-libvorbis --enable-libvo-amrwbenc --enable-libmysofa --enable-libspeex --enable-libxvid --enable-libaom --enable-libgsm --disable-w32threads --enable-libmfx --enable-ffnvcodec --enable-cuda-llvm --enable-cuvid --enable-d3d11va --enable-nvenc --enable-nvdec --enable-dxva2 --enable-avisynth --enable-libopenmpt --enable-amf
libavutil 56. 51.100 / 56. 51.100
libavcodec 58. 91.100 / 58. 91.100
libavformat 58. 45.100 / 58. 45.100
libavdevice 58. 10.100 / 58. 10.100
libavfilter 7. 85.100 / 7. 85.100
libswscale 5. 7.100 / 5. 7.100
libswresample 3. 7.100 / 3. 7.100
libpostproc 55. 7.100 / 55. 7.100

Answer 1

We may use loop filer for each frame, and concat the "looped" frames:
ffmpeg -r 2 -i img1.png -r 2 -i img2.png -r 2 -i img3.png -r 2 -i img4.png -i song1.mp3 -i song2.mp3 -i song3.wav -filter_complex "[4:a][5:a][6:a]concat=n=3:v=0:a=1[a];[0:v]scale=w=640:h=638,setsar=1,loop=505:505[v0];[1:v]scale=w=640:h=638,setsar=1,loop=505:505[v1];[2:v]scale=w=640:h=638,setsar=1,loop=505:505[v2];[3:v]scale=w=640:h=638,setsar=1,loop=505:505[v3];[v0][v1][v2][v3]concat=n=4:v=1:a=0,pad=ceil(iw/2)*2:ceil(ih/2)*2[v]" -map "[v]" -map "[a]" -c:a pcm_s32le -c:v libx264 -bufsize 3M -crf 18 -pix_fmt yuv420p -tune stillimage -t 1010 outputvideo.mkv
For making the command shorter, assume all the files are in the same folder.
-r 2 before each input image sets framerate of the input video to 2fps.
[4:a][5:a][6:a]concat=n=3:v=0:a=1[a] - concatenate the audio input files as in your question.
[0:v]scale=w=640:h=638,setsar=1 and [1:v]scale=w=640:h=638,setsar=1 ... - scales each one of the images to size 640x638 (assume the input images may have different sizes, we have to scale each image, before concatenating).
setsar=1 Makes sure that all the concatenated videos have the same SAR (Storage Aspect Ratio), and usually avoids aspect ratio deformation.
loop=505:505 - Uses loop filter for repeating each image 505 times (505 times at 2Hz applies 252.5 seconds).
The fist 505 is the number of loops, the second 505 is the number of frames in each loop (they should have the same value).
[0:v]scale=w=640:h=638,loop=505:505[v0] - applies scale filter, then loop 505 times, and store the result with temporary name [v0].
[v0][v1][v2][v3]concat=n=4:v=1:a=0 - concatenate the 4 videos (after scaling and looping).
pad=ceil(iw/2)*2:ceil(ih/2)*2[v] applies padding to the concatenated videos.
-map "[v]" -map "[a]" maps the video [v] to the output video channel, and the autio [a] to the output audio channel.
Since we have -t 1010, we don't need the -shortest argument (the output duration is defined to be 1010 seconds).