Best lossless audio working format for long runtime ?

[Gwar]

hello,
I'm about to do some re-dubbs of shows which run between 3-5 hours each.
Both audio formats of the original and the dub tracks
are in AAC 48000Hz stereo 161kbps which probably isn't a good working format. But when I convert theese files lossless
to uncompressed wav for example with a simple ffmpeg -acodec pcm_s32le -f WAV %1%.wav
They would be way too large to be recognised due to the filesize limit of the wav which is easily reached when using 24 or 32bit and having more than one channel and a longer runtime than 2 hours.
So would there be an alternative format I could convert this with ffmpeg to and then work with that in Davinci resolve and just export it as aac in the end.
It could be that I generate a loss when chosing aac as the final export format but I doubt that that would be audible I think its more important to process it and work with it lossless to preserve the quality.
thanks in advance for your help

[pandy]

Based on ffmpeg WAV muxer AVOptions:

  -write_bext        <boolean>    E.......... Write BEXT chunk. (default false)
  -write_peak        <int>        E.......... Write Peak Envelope chunk. (from 0 to 2) (default off)
     off             0            E.......... Do not write peak chunk.
     on              1            E.......... Append peak chunk after wav data.
     only            2            E.......... Write only peak chunk, omit wav data.
  -rf64              <int>        E.......... Use RF64 header rather than RIFF for large files. (from -1 to 1) (default never)
     auto            -1           E.......... Write RF64 header if file grows large enough.
     always          1            E.......... Always write RF64 header regardless of file size.
     never           0            E.......... Never write RF64 header regardless of file size.
  -peak_block_size   <int>        E.......... Number of audio samples used to generate each peak frame. (from 0 to 65536) (default 256)
  -peak_format       <int>        E.......... The format of the peak envelope data (1: uint8, 2: uint16). (from 1 to 2) (default 2)
  -peak_ppv          <int>        E.......... Number of peak points per peak value (1 or 2). (from 1 to 2) (default 2)

I would add '-rf64 -1' option to your commandline. Also instead integer i would be on favor to use float. So it will be like:

ffmpeg -c:a pcm_f32le -f WAV -rf64 -1 %1.wav

Also there can be some issue present in ffmpeg - check https://superuser.com/questions/1822472/how-to-convert-a-wav-to-wave-audio-ieee-float-...tereo-44100-hz

[Gwar]

that works awesome. thanks a lot