nepali_asr

Nepali Automatic Speech Recognition using pytorch

Project_major.ipynb is the file that should work.

usage:

 train.py   [-h] --save_model_path SAVE_MODEL_PATH
			[--load_model_from LOAD_MODEL_FROM]
	        [--resume_from_checkpoint RESUME_FROM_CHECKPOINT]
	        [--load_x LOAD_X] [--load_y LOAD_Y] [--logdir LOGDIR]

optional arguments:

	-h, --help show this help message and exit

		--save_model_path SAVE_MODEL_PATH

	path to save model

		--load_model_from LOAD_MODEL_FROM

	path to load a pretrain model to continue training

		--resume_from_checkpoint RESUME_FROM_CHECKPOINT

	check path to resume from

		--load_x LOAD_X path to load a tensor x file
		--load_y LOAD_Y path to load a tensor label file
		--logdir LOGDIR path to save logs