lrenault/ddsp-piano

MIDI Piano synthesizer using DDSP.

ddsp-piano's Issues

ModuleNotFoundError: No module named 'priv_ddfx'

Thank you for your research and sharing. I tried with the main repo and there was no problem, but when I tried with the dev repo I got the error: ModuleNotFoundError: No module named 'priv_ddfx'
I think your repo is missing module priv_ddfx.py?

Hi, is it possible to train at a higher sample rate?

Help please. ValueError: operands could not be broadcast together with remapped shapes

Thanks for sharing your codes and models! I tried to run 'synthesize_midi_file.py' as you said, but encountered the below error:

(ddsp_test) F:\code\ddsp-piano-dev>python synthesize_midi_file.py "test01.mid" "test01.wav"
Loading midi file...
Traceback (most recent call last):
File "synthesize_midi_file.py", line 63, in
main(process_args())
File "synthesize_midi_file.py", line 32, in main
inputs = load_midi_as_conditioning(args.midi_file, duration=args.duration)
File "F:\code\ddsp-piano-dev\ddsp_piano\utils\io_utils.py", line 102, in load_midi_as_conditioning
conditioning = ensure_sequence_length(conditioning, target_n_frames)
File "F:\code\ddsp-piano-dev\ddsp_piano\utils\io_utils.py", line 61, in ensure_sequence_length
return np.pad(sequence, pad_width=pad_width)
File "<array_function internals>", line 6, in pad
File "F:\Envs\ddsp_test\lib\site-packages\numpy\lib\arraypad.py", line 743, in pad
pad_width = _as_pairs(pad_width, array.ndim, as_index=True)
File "F:\Envs\ddsp_test\lib\site-packages\numpy\lib\arraypad.py", line 518, in _as_pairs
return np.broadcast_to(x, (ndim, 2)).tolist()
File "<array_function internals>", line 6, in broadcast_to
File "F:\Envs\ddsp_test\lib\site-packages\numpy\lib\stride_tricks.py", line 411, in broadcast_to
return _broadcast_to(array, shape, subok=subok, readonly=True)
File "F:\Envs\ddsp_test\lib\site-packages\numpy\lib\stride_tricks.py", line 350, in _broadcast_to
op_flags=['readonly'], itershape=shape, order='C')
ValueError: operands could not be broadcast together with remapped shapes [original->remapped]: (6,) and requested shape (3,2)