44khz, issues?

[Rose-sys]

After trying to go 44khz, I am getting errors trying to convert:

Traceback (most recent call last):
  File "scripts/voice_change.py", line 79, in process
    f_out = acoustic_converter.combine_silent(effective=effective, feature=f_out)
  File "/home/localuser/.local/lib/python3.7/site-packages/yukarin/acoustic_converter.py", line 162, in combine_silent
    silent_feature.indexing_set(effective, feature)
  File "/home/localuser/.local/lib/python3.7/site-packages/yukarin/acoustic_feature.py", line 100, in indexing_set
    self.ap[index] = feature.ap
ValueError: shape mismatch: value array of shape (1013,513) could not be broadcast to indexing result of shape (1013,1025)

I redid all steps from beginning using --sampling-rate 44000 however I am getting these errors.

Now I wonder if I am doing anything wrong.

python3 scripts/voice_change.py \
	--model_dir './model_stage1' \
	--config_path './model_stage1/config.json' \
	--input_statistics '../../voice/input_statistics.npy' \ 
	--target_statistics '../../voice/target_statistics.npy' \
	--output_sampling_rate 44000 \
	--disable_dataset_test \ 
	--test_wave_dir '../../voice/testwave/' \
	--output_dir '../../voice/outputtest/' 

in the sample_config.json as well config.json the value is set:
"sampling_rate": 44000