Skeleton framework for item generation scripts (323e4b86) · Commits · IVAS Codec Public Collaboration / IVAS Processing Scripts

item_generation_scripts/init.py

0 → 100644

+101 −0

Original line number	Diff line number	Diff line
		#!/usr/bin/env python3

		#
		# (C) 2022-2023 IVAS codec Public Collaboration with portions copyright Dolby International AB, Ericsson AB,
		# Fraunhofer-Gesellschaft zur Foerderung der angewandten Forschung e.V., Huawei Technologies Co. LTD.,
		# Koninklijke Philips N.V., Nippon Telegraph and Telephone Corporation, Nokia Technologies Oy, Orange,
		# Panasonic Holdings Corporation, Qualcomm Technologies, Inc., VoiceAge Corporation, and other
		# contributors to this repository. All Rights Reserved.
		#
		# This software is protected by copyright law and by international treaties.
		# The IVAS codec Public Collaboration consisting of Dolby International AB, Ericsson AB,
		# Fraunhofer-Gesellschaft zur Foerderung der angewandten Forschung e.V., Huawei Technologies Co. LTD.,
		# Koninklijke Philips N.V., Nippon Telegraph and Telephone Corporation, Nokia Technologies Oy, Orange,
		# Panasonic Holdings Corporation, Qualcomm Technologies, Inc., VoiceAge Corporation, and other
		# contributors to this repository retain full ownership rights in their respective contributions in
		# the software. This notice grants no license of any kind, including but not limited to patent
		# license, nor is any license granted by implication, estoppel or otherwise.
		#
		# Contributors are required to enter into the IVAS codec Public Collaboration agreement before making
		# contributions.
		#
		# This software is provided "AS IS", without any express or implied warranties. The software is in the
		# development stage. It is intended exclusively for experts who have experience with such software and
		# solely for the purpose of inspection. All implied warranties of non-infringement, merchantability
		# and fitness for a particular purpose are hereby disclaimed and excluded.
		#
		# Any dispute, controversy or claim arising under or in relation to providing this software shall be
		# submitted to and settled by the final, binding jurisdiction of the courts of Munich, Germany in
		# accordance with the laws of the Federal Republic of Germany excluding its conflict of law rules and
		# the United Nations Convention on Contracts on the International Sales of Goods.
		#

		import os
		import logging
		from itertools import repeat
		import yaml
		import pdb

		from item_generation_scripts.constants import (
		LOGGER_DATEFMT,
		LOGGER_FORMAT,
		LOGGER_SUFFIX,
		)
		from item_generation_scripts.processing import config, process_ism_items
		from item_generation_scripts.processing import config
		from item_generation_scripts.utils import create_dir


		def logging_init(args, cfg):
		"""set up logging for a test file"""
		logger = logging.getLogger("__main__")
		logger.setLevel(logging.DEBUG)

		# console handler
		console_handler = logging.StreamHandler()
		console_handler.setFormatter(logging.Formatter("%(message)s"))
		console_handler.setLevel(logging.DEBUG if args.debug else logging.INFO)
		logger.addHandler(console_handler)

		# main log file
		file_handler = logging.FileHandler(
		cfg.output_path.joinpath(f"{cfg.format}{LOGGER_SUFFIX}"), mode="w"
		)
		file_handler.setFormatter(logging.Formatter(LOGGER_FORMAT, datefmt=LOGGER_DATEFMT))
		file_handler.setLevel(logging.DEBUG if args.debug else logging.INFO)
		logger.addHandler(file_handler)

		logger.info(f"Processing test configuration file {args.config}")
		logger.info(f"Input path: {cfg.input_path.absolute()}")
		logger.info(f"Output path: {cfg.output_path.absolute()}")

		return logger


		def main(args):

		# parse configuration
		cfg = config.TestConfig(args.config)

		# create output directory, if not existing
		if not os.path.exists(cfg.output_path):
		create_dir(cfg.output_path)

		# set up logging
		logger = logging_init(args, cfg)

		# generate input items
		if cfg.format.startswith("ISM"):
		# generate ISM items according to scene description
		process_ism_items.generate_ism_items(
		cfg.format,
		cfg.loudness,
		cfg.input_path,
		cfg.output_path,
		cfg.scenes,
		logger
		)

		# copy configuration to output directory
		with open(cfg.output_path.joinpath(f"{cfg.format}.yml"), "w") as f:
		yaml.safe_dump(cfg._yaml_dump, f)

item_generation_scripts/main.py

0 → 100644

+50 −0

Original line number	Diff line number	Diff line
		#!/usr/bin/env python3

		#
		# (C) 2022-2023 IVAS codec Public Collaboration with portions copyright Dolby International AB, Ericsson AB,
		# Fraunhofer-Gesellschaft zur Foerderung der angewandten Forschung e.V., Huawei Technologies Co. LTD.,
		# Koninklijke Philips N.V., Nippon Telegraph and Telephone Corporation, Nokia Technologies Oy, Orange,
		# Panasonic Holdings Corporation, Qualcomm Technologies, Inc., VoiceAge Corporation, and other
		# contributors to this repository. All Rights Reserved.
		#
		# This software is protected by copyright law and by international treaties.
		# The IVAS codec Public Collaboration consisting of Dolby International AB, Ericsson AB,
		# Fraunhofer-Gesellschaft zur Foerderung der angewandten Forschung e.V., Huawei Technologies Co. LTD.,
		# Koninklijke Philips N.V., Nippon Telegraph and Telephone Corporation, Nokia Technologies Oy, Orange,
		# Panasonic Holdings Corporation, Qualcomm Technologies, Inc., VoiceAge Corporation, and other
		# contributors to this repository retain full ownership rights in their respective contributions in
		# the software. This notice grants no license of any kind, including but not limited to patent
		# license, nor is any license granted by implication, estoppel or otherwise.
		#
		# Contributors are required to enter into the IVAS codec Public Collaboration agreement before making
		# contributions.
		#
		# This software is provided "AS IS", without any express or implied warranties. The software is in the
		# development stage. It is intended exclusively for experts who have experience with such software and
		# solely for the purpose of inspection. All implied warranties of non-infringement, merchantability
		# and fitness for a particular purpose are hereby disclaimed and excluded.
		#
		# Any dispute, controversy or claim arising under or in relation to providing this software shall be
		# submitted to and settled by the final, binding jurisdiction of the courts of Munich, Germany in
		# accordance with the laws of the Federal Republic of Germany excluding its conflict of law rules and
		# the United Nations Convention on Contracts on the International Sales of Goods.
		#

		import argparse

		from item_generation_scripts import main

		if __name__ == "__main__":
		parser = argparse.ArgumentParser(
		description="IVAS item generation scripts for listening tests. Please refer to README.md for usage."
		)
		parser.add_argument(
		"config",
		help="YAML configuration file",
		)
		parser.add_argument(
		"--debug", help="Set logging level to debug", action="store_true", default=False
		)
		args = parser.parse_args()

		main(args)

item_generation_scripts/audiotools/EFAP.py

0 → 100644

+922 −0

File added.

Preview size limit exceeded, changes collapsed.

item_generation_scripts/audiotools/init.py

0 → 100644

+286 −0

Original line number	Diff line number	Diff line
		#!/usr/bin/env python3

		#
		# (C) 2022-2023 IVAS codec Public Collaboration with portions copyright Dolby International AB, Ericsson AB,
		# Fraunhofer-Gesellschaft zur Foerderung der angewandten Forschung e.V., Huawei Technologies Co. LTD.,
		# Koninklijke Philips N.V., Nippon Telegraph and Telephone Corporation, Nokia Technologies Oy, Orange,
		# Panasonic Holdings Corporation, Qualcomm Technologies, Inc., VoiceAge Corporation, and other
		# contributors to this repository. All Rights Reserved.
		#
		# This software is protected by copyright law and by international treaties.
		# The IVAS codec Public Collaboration consisting of Dolby International AB, Ericsson AB,
		# Fraunhofer-Gesellschaft zur Foerderung der angewandten Forschung e.V., Huawei Technologies Co. LTD.,
		# Koninklijke Philips N.V., Nippon Telegraph and Telephone Corporation, Nokia Technologies Oy, Orange,
		# Panasonic Holdings Corporation, Qualcomm Technologies, Inc., VoiceAge Corporation, and other
		# contributors to this repository retain full ownership rights in their respective contributions in
		# the software. This notice grants no license of any kind, including but not limited to patent
		# license, nor is any license granted by implication, estoppel or otherwise.
		#
		# Contributors are required to enter into the IVAS codec Public Collaboration agreement before making
		# contributions.
		#
		# This software is provided "AS IS", without any express or implied warranties. The software is in the
		# development stage. It is intended exclusively for experts who have experience with such software and
		# solely for the purpose of inspection. All implied warranties of non-infringement, merchantability
		# and fitness for a particular purpose are hereby disclaimed and excluded.
		#
		# Any dispute, controversy or claim arising under or in relation to providing this software shall be
		# submitted to and settled by the final, binding jurisdiction of the courts of Munich, Germany in
		# accordance with the laws of the Federal Republic of Germany excluding its conflict of law rules and
		# the United Nations Convention on Contracts on the International Sales of Goods.
		#

		import argparse
		from itertools import repeat
		from pathlib import Path

		from item_generation_scripts.audiotools.constants import AUDIO_FORMATS
		from item_generation_scripts.audiotools.convert import convert_file
		from item_generation_scripts.utils import apply_func_parallel


		def add_processing_args(group, input=True):
		# set up prefixes to avoid argument collision
		if input:
		p = "in"
		ps = "i"
		else:
		p = "out"
		ps = "o"

		group.add_argument(
		f"-{ps}",
		f"--{p}",
		dest=f"{p}put",
		required=True,
		type=Path,
		help="Path to *.{wav, pcm, raw} file or directory",
		)
		group.add_argument(
		f"-{ps}f",
		f"--{p}_fmt",
		required=input,
		type=str,
		help="Audio format (use -l, --list for a list / -L, --long for a detailed list)",
		default=None,
		)
		group.add_argument(
		f"-{ps}s",
		f"--{p}_fs",
		type=int,
		help="Sampling rate (Hz) (deduced for .wav input, same as input if output not specified, default = %(default)s)",
		default=48000,
		)
		group.add_argument(
		f"-{ps}fc",
		f"--{p}_cutoff",
		type=int,
		help="Cut-off frequency for low-pass filtering (default = %(default)s)",
		default=None,
		)
		group.add_argument(
		f"-{ps}hp",
		f"--{p}_hp50",
		help="Apply 50 Hz high-pass filtering (default = %(default)s)",
		action="store_true",
		)
		group.add_argument(
		f"-{ps}w",
		f"--{p}_window",
		type=float,
		help="Window the start/end of the signal by this amount in milliseconds (default = %(default)s)",
		default=None,
		)
		group.add_argument(
		f"-{ps}t",
		f"--{p}_trim",
		type=float,
		nargs=2,
		metavar=("PRE_TRIM", "POST_TRIM"),
		help="Pre-/post-trim the signal by this amount in milliseconds (negative values pad silence), (default = %(default)s)",
		)
		group.add_argument(
		f"-{ps}pn",
		f"--{p}_pad_noise",
		help="Flag for padding with noise instead of zeros",
		action="store_true",
		)
		group.add_argument(
		f"-{ps}d",
		f"--{p}_delay",
		type=float,
		help="Delay the signal by this amount in milliseconds (negative values advance, default = %(default)s)",
		default=None,
		)
		group.add_argument(
		f"-{ps}l",
		f"--{p}_loudness",
		type=float,
		help="Normalize to given loudness with BS 1770-4 (default = %(default)s)",
		default=None,
		)
		group.add_argument(
		f"-{ps}nf",
		f"--{p}_loudness_fmt",
		type=str,
		help=f"Format used for loudness computation (only valid with with -{ps}l/--{p}_loudness, default = {p.upper()}_FMT)",
		default=None,
		)


		def get_args():
		parser = argparse.ArgumentParser(
		description="Audiotools: Convert/Manipulate spatial audio files."
		)

		""" Input file arguments """
		input_parser = parser.add_argument_group("Input (pre-) processing options")

		# add common arguments
		add_processing_args(input_parser)

		# input only arguments
		input_parser.add_argument(
		"-im",
		"--in_meta",
		type=str,
		nargs="+",
		help="list of input metadata files (only relevant for ISM and MASA input)",
		default=None,
		)

		""" Output file arguments """
		output_parser = parser.add_argument_group("Output (post-) processing options")

		# add common arguments
		add_processing_args(output_parser, False)

		# output only arguments
		output_parser.add_argument(
		"-lm",
		"--limit",
		help="Apply limiting to output (default = %(default)s)",
		action="store_true",
		)
		output_parser.add_argument(
		"-t",
		"--trajectory",
		type=str,
		help="Head-tracking trajectory file for binaural output (default = %(default)s)",
		default=None,
		)
		output_parser.add_argument(
		"-bd",
		"--bin_dataset",
		type=str,
		help="Use a custom binaural dataset (see README.md and audiotools/binaural_datasets/README.txt for further information)",
		default=None,
		)
		output_parser.add_argument(
		"-bl",
		"--bin_lfe_gain",
		type=float,
		help="Render LFE to binaural output with the specified gain (only valid for channel-based input, default = %(default)s)",
		default=None,
		)
		output_parser.add_argument(
		"-mnru",
		"--mnru_q",
		type=float,
		help="Flag for MNRU processing",
		default=None,
		)
		output_parser.add_argument(
		"-esdru",
		"--esdru_alpha",
		type=float,
		help="Flag for ESDRU processing",
		default=None,
		)

		misc_parser = parser.add_argument_group("General options")

		""" Miscellaneous or meta arguments """
		misc_parser.add_argument(
		"-l",
		"--list",
		help="list all supported audio formats and exit",
		action="store_true",
		)
		misc_parser.add_argument(
		"-L",
		"--long",
		help="list all supported audio formats with long description and exit",
		action="store_true",
		)
		misc_parser.add_argument(
		"-mp",
		"--multiprocessing",
		help="Enable multiprocessing (default = %(default)s)",
		action="store_true",
		)

		return parser.parse_args()


		def main():
		args = get_args()

		if args.list is True or args.long is True:
		for fmt in AUDIO_FORMATS:
		if args.long:
		for f, d in fmt.items():
		print(f)
		[print(f"\t{k}: {v}", end=None) for k, v in d.items()]
		else:
		print(", ".join(fmt.keys()))
		exit()

		elif args.input is not None:
		if not args.out_fs:
		args.out_fs = args.in_fs

		if not args.out_fmt:
		args.out_fmt = args.in_fmt

		if not args.out_loudness_fmt:
		args.out_loudness_fmt = args.out_fmt

		# List input files
		args.input = Path(args.input)
		in_files = []
		if args.input.exists():
		if args.input.is_dir():
		in_files.extend(args.input.glob("*.wav"))
		in_files.extend(args.input.glob("*.pcm"))
		in_files.extend(args.input.glob("*.raw"))
		else:
		in_files = [args.input]
		else:
		raise ValueError(f"Input path {args.input} does not exist!")

		if len(in_files) == 0:
		raise ValueError(f"Input directory {args.input} empty!")

		# Create output directory
		args.output = Path(args.output)

		if len(in_files) == 1 and args.input.is_file():
		out_files = [args.output]
		else:
		args.output.mkdir(exist_ok=True)
		out_files = [args.output.joinpath(i.name) for i in in_files]

		# Multiprocessing
		enable_multiprocessing = args.multiprocessing

		# Remove unneeded keys to avoid passing to convert_file()
		for k in ["list", "long", "multiprocessing", "input", "output"]:
		args.__dict__.pop(k)

		apply_func_parallel(
		convert_file,
		zip(in_files, out_files),
		repeat(args.__dict__),
		"mp" if enable_multiprocessing else None,
		)

item_generation_scripts/audiotools/main.py

0 → 100644

+36 −0

Original line number	Diff line number	Diff line
		#!/usr/bin/env python3

		#
		# (C) 2022-2023 IVAS codec Public Collaboration with portions copyright Dolby International AB, Ericsson AB,
		# Fraunhofer-Gesellschaft zur Foerderung der angewandten Forschung e.V., Huawei Technologies Co. LTD.,
		# Koninklijke Philips N.V., Nippon Telegraph and Telephone Corporation, Nokia Technologies Oy, Orange,
		# Panasonic Holdings Corporation, Qualcomm Technologies, Inc., VoiceAge Corporation, and other
		# contributors to this repository. All Rights Reserved.
		#
		# This software is protected by copyright law and by international treaties.
		# The IVAS codec Public Collaboration consisting of Dolby International AB, Ericsson AB,
		# Fraunhofer-Gesellschaft zur Foerderung der angewandten Forschung e.V., Huawei Technologies Co. LTD.,
		# Koninklijke Philips N.V., Nippon Telegraph and Telephone Corporation, Nokia Technologies Oy, Orange,
		# Panasonic Holdings Corporation, Qualcomm Technologies, Inc., VoiceAge Corporation, and other
		# contributors to this repository retain full ownership rights in their respective contributions in
		# the software. This notice grants no license of any kind, including but not limited to patent
		# license, nor is any license granted by implication, estoppel or otherwise.
		#
		# Contributors are required to enter into the IVAS codec Public Collaboration agreement before making
		# contributions.
		#
		# This software is provided "AS IS", without any express or implied warranties. The software is in the
		# development stage. It is intended exclusively for experts who have experience with such software and
		# solely for the purpose of inspection. All implied warranties of non-infringement, merchantability
		# and fitness for a particular purpose are hereby disclaimed and excluded.
		#
		# Any dispute, controversy or claim arising under or in relation to providing this software shall be
		# submitted to and settled by the final, binding jurisdiction of the courts of Munich, Germany in
		# accordance with the laws of the Federal Republic of Germany excluding its conflict of law rules and
		# the United Nations Convention on Contracts on the International Sales of Goods.
		#

		from item_generation_scripts.audiotools import main

		if __name__ == "__main__":
		main()