Workflow for recording a talk and post-processing it

Capture screen and in-build notebook microphone during talk with OBS
Cut the video to the desired length with Kdenlive. Export as h264 for maximum end-user compatibility
Upload recording to auphonic.com for automated processing (volume leveling -> pretty clean audio, speech recognition -> subtitles)
Correct generated subtitles with Subtitle Composer (latest git development version for WebVTT (.vtt) support)
Export subtitles as SubRip (.srt) with Suptitle Composer
Translate the .srt with deepl.com by opening the file with a text editor and copying twenty captions at a time for translation (auto translation from Subtitle Composer was not working for me). Save the translations in a new .srt
Embed the subtitles with ffmpeg with the following command (taken from bannerbear.com)

ffmpeg -i input.mp4 -i subtitle.en.srt -i subtitle.chi.srt -map 0 -map 1 -map 2 -c copy -c:s mov_text -metadata:s:s:0 language=eng -metadata:s:s:1 language=chi output_eng_chi.mp4
Share :)

Blog