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.
9 lines
445 B
Plaintext
9 lines
445 B
Plaintext
# Additive on top of the pytorch/pytorch base image's own CUDA-matched torch
|
|
# build (see Dockerfile) - scipy is needed only because models/utils.py has a
|
|
# top-level `from scipy.io.wavfile import read` import; Cython is needed only
|
|
# to build the monotonic_align extension (imported at module load by
|
|
# models/models.py even though this sidecar's infer()-only path never calls
|
|
# it - see the comment in Dockerfile).
|
|
scipy==1.16.3
|
|
Cython>=3.0.0
|