Many (perhaps the majority) of snipped audio files have an artifact in the very beginning. This StackExchange thread mentions a similar issue (albeit in a much different application). Is this an ffmpeg problem? Can it be avoided? Can we use Audacity batch processing instead of ffmpeg?
Closing this issue; it actually seems to be a (perfectly reproducible) artifact of the Ubuntu video player (Totem) and does not occur in VLC or Audacity.
Many (perhaps the majority) of snipped audio files have an artifact in the very beginning. This StackExchange thread mentions a similar issue (albeit in a much different application). Is this an ffmpeg problem? Can it be avoided? Can we use Audacity batch processing instead of ffmpeg?