28 lines
1.2 KiB
Python
28 lines
1.2 KiB
Python
# Copyright 2013-2023 Lawrence Livermore National Security, LLC and other
|
|
# Spack Project Developers. See the top-level COPYRIGHT file for details.
|
|
#
|
|
# SPDX-License-Identifier: (Apache-2.0 OR MIT)
|
|
|
|
from spack.package import *
|
|
|
|
|
|
class PyNaraWpe(PythonPackage):
|
|
"""Background noise and signal reverberation due to reflections in an
|
|
enclosure are the two main impairments in acoustic signal processing and
|
|
far-field speech recognition. This work addresses signal dereverberation
|
|
techniques based on WPE for speech recognition and other far-field
|
|
applications. WPE is a compelling algorithm to blindly dereverberate
|
|
acoustic signals based on long-term linear prediction."""
|
|
|
|
homepage = "https://github.com/fgnt/nara_wpe"
|
|
pypi = "nara_wpe/nara_wpe-0.0.7.tar.gz"
|
|
|
|
version("0.0.7", sha256="7aa2edd5261e5d953e584e69a9233d60fc588fc8a4b7886c3ce43cc8ac8cd99b")
|
|
|
|
depends_on("py-setuptools", type="build")
|
|
depends_on("py-numpy", type=("build", "run"))
|
|
depends_on("py-tqdm", type=("build", "run"))
|
|
depends_on("py-soundfile", type=("build", "run"))
|
|
depends_on("py-bottleneck", type=("build", "run"))
|
|
depends_on("py-click", type=("build", "run"))
|