tsukumi 15af441cba Improve: Convert ONNX version of BERT models to FP16 ...

I have found that half-precision has little effect on speech synthesis quality and, depending on the environment, can reduce file size and memory usage by half, so I have decided to use FP16.

2024-12-19 14:51:30 +09:00

..

models

Fix: errors during ONNX CPU inference

2024-10-06 16:18:21 +09:00

nlp

Improve: Convert ONNX version of BERT models to FP16

2024-12-19 14:51:30 +09:00

utils

Add: test code for CoreMLExecutionProvider on a trial basis

2024-09-23 21:07:27 +09:00

__init__.py

Refactor: rewrote Japanese natural language processing code imported from server_editor.py

2024-03-06 20:56:21 +00:00

.editorconfig

Refactor: rewrote Japanese natural language processing code imported from server_editor.py

2024-03-06 20:56:21 +00:00

constants.py

Bump ver only

2024-11-11 10:18:55 +09:00

logging.py

Refactor

2024-03-15 12:27:46 +09:00

tts_model.py

Fix: TTSModel.unload() did not work in PyTorch-independent environments

2024-12-19 13:51:52 +09:00

voice.py

docs

2024-03-13 15:46:02 +09:00