IOError: [Errno 2] No such file or directory: 'database/feats/f0/wan_a.f0_world'

[syazw]

wan_a This function cannot support stereo file error: The file is not .wav format. Running REAPER f0 extraction... sh: 1: /home/zainursyazwana/audio_wan/merlin/tools/bin/REAPER/reaper: not found Traceback (most recent call last): File "/home/zainursyazwana/audio_wan/merlin/misc/scripts/vocoder/world/extract_features_for_merlin.py", line 216, in pool.map(process, wav_files) File "/usr/lib/python2.7/multiprocessing/pool.py", line 251, in map return self.map_async(func, iterable, chunksize).get() File "/usr/lib/python2.7/multiprocessing/pool.py", line 567, in get raise self._value IOError: [Errno 2] No such file or directory: 'database/feats/f0/wan_a.f0_world' Copying features to acoustic data directory... done...! ##################################################################

Hi @ronanki , I got this error, when i tried to run ./03_prepare_acoustic_features.sh database/wav database/feats . So, how to fix this error? Thanks for your help.

[felipeespic]

Hi @syazw , Did you installed REAPER? Also, if your files are stereo, convert them to mono before extracting features.

[HandsomeDevilv112]

~where was the installation file/instructions for REAPER?~ in compile_tools I have no idea how I managed to avoid it.

[Jackiexiao]

@HandsomeDevilv112 if you install REAPER successfully, you could find REAPER in merlin/tools/REAPER, otherwise, see merlin/tools/compile_tools.sh which install REAPER as well as other essential tools like htk world sptk

[HandsomeDevilv112]

@Jackiexiao it's been a number of months, but I just know I did compile_tools/compile_other_tools/compile_htk/compile_compile_compile. Apologies, friend. I have periodic fits of illiteracy.

[Maxxiey]

Hi, @HandsomeDevilv112 @Jackiexiao , I got the same error here, I checked the REAPER directory and all seemed to be normal. I successfully trained merlin on a little dataset, but when I tried to use a bigger one(more than 10000), the error was raised.

How did you guys solve that problem? thanks in advance.

[RasmusD]

@Maxxiey If youre error is the exact same as the initially reported error. Then at least 1 of your wav files is a stereo file. Probably one in the larger dataset. Convert it to mono.

[Maxxiey]

@RasmusD Thanks and sorry for the late reply, I checked out the log and find that there were error messages like '**.wav is not wav format'. I searched the issue list and pinpointed that the error was raised because some of the wav files contained metadata, and the script can not recognise those wav files. I used ffmpeg to remove the metadata part and everything went right.

And yes, if the wav files are stereo, there will be problems, need to convert them first~