I'm using the following ffmpeg command to convert a multitrack wav to multiple mono tracks:
ffmpeg -i input.wav \ -map_channel 0.0.0 a.wav \ -map_channel 0.0.1 b.wav \ -map_channel 0.0.2 c.wav \ -map_channel 0.0.3 d.wav \This works fine but it converts the channels to 16bit PCM:
...
Stream #0:0 -> #0:0 (pcm_s24le (native) -> pcm_s16le (native))
Stream #0:0 -> #1:0 (pcm_s24le (native) -> pcm_s16le (native))
Stream #0:0 -> #2:0 (pcm_s24le (native) -> pcm_s16le (native))
Stream #0:0 -> #3:0 (pcm_s24le (native) -> pcm_s16le (native))
...I've also tried "-sample_fmt pcm_s24le" or "-sample_fmt s32" but both parameters were not recognized.
How can I extract the channels of a multiwav file without changing the bit depth?
1 Answer
The encoder pcm_s16le is being used for your .wav output. This encoder only supports the s16 sample format (see ffmpeg -h encoder=pcm_s16le and ffmpeg -sample_fmts).
You can choose the encoder pcm_s24le:
ffmpeg -i input.wav \ -map_channel 0.0.0 -c:a pcm_s24le a.wav \ -map_channel 0.0.1 -c:a pcm_s24le b.wav \ -map_channel 0.0.2 -c:a pcm_s24le c.wav \ -map_channel 0.0.3 -c:a pcm_s24le d.wav
