Issues with make composition of acoustic features

[Lyu T]

When I tried to make composition of my own acoustic features, I got this problem:

File “/home/lvtongtong/dnn_tts/run_dnn.py”, line 970, in <module>
main_function(cfg)
File “/home/lvtongtong/dnn_tts/run_dnn.py”, line 635, in main_function
acoustic_worker.prepare_nn_data(in_file_list_dict, nn_cmp_file_list, cfg.in_dimension_dict, cfg.out_dimension_dict)
File “/home/lvtongtong/dnn_tts/frontend/acoustic_base.py”, line 122, in prepare_nn_data
self.prepare_data(in_file_list_dict, out_file_list, in_dimension_dict, out_dimension_dict)
File “/home/lvtongtong/dnn_tts/frontend/acoustic_composition.py”, line 126, in prepare_data
features, frame_number = io_funcs.load_binary_file_frame(in_file_name, in_feature_dim)
File “/home/lvtongtong/dnn_tts/io_funcs/binary_io.py”, line 67, in load_binary_file_frame
assert features.size % float(dimension) == 0.0,’specified dimension %s not compatible with data’%(dimension)
AssertionError: specified dimension 5 not compatible with data.

Is there anything wrong in the step of generating acoustic features? My data is from a small Chinese data set.

Thanks for your time.

[Srikanth R]

The dimension of your acoustic feature is not 5. Check which features, you are using and it’s dimension.

[Lyu T]

It should be the problem of bap because I set the dimension of bap as 5. Should I change it as 1 and does it influence the following steps? Thanks.

[Author]

Posts