Nik Afiq
5238298b55
- Implemented core model components in `modules.py` including various convolutional layers and normalization techniques. - Added transformation functions in `transforms.py` for piecewise rational quadratic transformations. - Created utility functions in `utils.py` for checkpoint management, logging, and hyperparameter handling. - Introduced monotonic alignment functionality with Cython optimization in `monotonic_align`. - Developed a minimal inference server in `server.py` to handle synthesis requests. - Updated requirements to include necessary dependencies for Cython and scipy.
20 lines
612 B
Python
20 lines
612 B
Python
import numpy as np
|
|
import torch
|
|
from .monotonic_align.core import maximum_path_c
|
|
|
|
|
|
def maximum_path(neg_cent, mask):
|
|
""" Cython optimized version.
|
|
neg_cent: [b, t_t, t_s]
|
|
mask: [b, t_t, t_s]
|
|
"""
|
|
device = neg_cent.device
|
|
dtype = neg_cent.dtype
|
|
neg_cent = neg_cent.data.cpu().numpy().astype(np.float32)
|
|
path = np.zeros(neg_cent.shape, dtype=np.int32)
|
|
|
|
t_t_max = mask.sum(1)[:, 0].data.cpu().numpy().astype(np.int32)
|
|
t_s_max = mask.sum(2)[:, 0].data.cpu().numpy().astype(np.int32)
|
|
maximum_path_c(path, neg_cent, t_t_max, t_s_max)
|
|
return torch.from_numpy(path).to(device=device, dtype=dtype)
|