Commit 71a48872 authored by Anika Treffehn's avatar Anika Treffehn
Browse files

reference renderer and loudness tool updates

parent eb7d0087

scripts/README.md

+3 −3
Original line number Diff line number Diff line
@@ -493,17 +493,17 @@ This is the main file to edit in order to change global configuration options, d 
|                           |                    |               |                                               |

| concatenate_input         | True/False         | False         | Whether to (horizontally) concatenate files in the input directory |

| concat_silence_ms         | [1000, 1000]       | [0, 0]        | Specifies the pre- and post-silence duration to pad concatenation with in ms. If a single value is specified it will be used for BOTH pre- and post-padding |

| preproc_loudness          | -26                |               | Loudness to preprocess input to (dBov / LKFS depending on tool). Only processed if preproc_input is True. |

| preproc_loudness          | -26                |               | Loudness to preprocess input to (in LKFS). Only processed if preproc_input is True. |

|                           |                    |               |                                               |

| output_path               | ./out/             |               | Output root directory hosting generated items & log |

| out_fs                    | 48000              | 48000         | Output sampling rate for conditions to generate |

| output_loudness           | -26                |               | Loudness level for output file (dBov / LKFS depending on tool). |

| output_loudness           | -26                |               | Loudness level for output file in LKFS. |

|                           |                    |               |                                               |

| renderer_format           | 7_1_4 or CICP19    | Required      | Format to be rendered (using offline rendering, will be bypassed if = out_format) |

| binaural_rendered         | True/False         | False         | Extra binauralization of the rendered outputs (using offline rendering) |

| include_LFE               | True/False         | False         | Whether to include LFE in binural rendering   |

| gain_factor               | float value        | 1.0           | Gain factor to be applied to LFE channel      |

| loudness_tool             | "sv56demo"         | "bs1770demo"  | Tool to use for loudness adjustment. Currently only sv56demo and bs1770demo are supported for appropriate format configurations. Optionally can be a path to the binary.  |

| loudness_format           | "scale_out"/CICP19 | renderer_format | Rendering format used in iterative loudness adjustment. "scale_out" just scales the output to the desired output_loudness |  |

|                           |                    |               |                                               |

| lt_mode                   | "MUSHRA"           |               | Automatically generates a NAME.ltg file with generate_lt_file.py in output_path according to the specified mode |

| conditions_to_generate    | ["ref", "ivas"]    | Required      | list of conditions to be generated, for ivas and evs, multiple conditions can be specified with an \_ separator (i.e. "ivas_branch", "ivas_trunk" etc.)            |

scripts/generate_test_items.py

+9 −0
Original line number Diff line number Diff line
@@ -36,6 +36,7 @@ import multiprocessing as mp 
import os

from typing import Iterable



from generate_lt_file import generate_lt_file

from pyaudio3dtools import audiofile

from pyprocessing import processing, processing_configs, utils
@@ -171,6 +172,14 @@ def main(test_cfg): 
        for r in results:

            r.get()



    # generate LT file if specified

    if hasattr(test_cfg, "lt_mode"):

        generate_lt_file(

            test_cfg,

            test_cfg.lt_mode,

            os.path.join(test_cfg.output_path, test_cfg.name + ".ltg"),

        )



    # copy over JSON to main output directory

    output_json = os.path.join(test_cfg.output_path, test_cfg.name + ".json")

    with open(output_json, "w") as fp:

scripts/pyaudio3dtools/EFAP.py

+22 −31
Original line number Diff line number Diff line 
#!/usr/bin/env python3



"""

   (C) 2022 Baseline Development Group with portions copyright Dolby International AB, Ericsson AB,

   (C) 2022-2023 Baseline Development Group 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 Corporation, Qualcomm Technologies, Inc., VoiceAge Corporation. All Rights Reserved.
@@ -36,7 +36,9 @@ import numpy as np 




def wrap_angles(

    azi: float, ele: float, clip_ele: Optional[bool] = False

    azi: float,

    ele: float,

    clip_ele: Optional[bool] = False,

) -> Tuple[float, float]:

    """

    Wrap angles to (-180, 180] azimuth and [-90, 90] elevation
@@ -45,7 +47,7 @@ def wrap_angles( 
    if clip_ele:

        ele = min(max(ele, -90), 90)



    if ele != 0 and ele % 90 == 0:

    if ele % 90 == 0 and ele % 180 != 0:

        # if elevation is a multiple of 90, azimuth is irrelevant since we are at a pole

        azi = 0

        while np.abs(ele) > 90:
@@ -58,20 +60,15 @@ def wrap_angles( 


            # compensate elevation accordingly

            if ele > 90:

                ele -= 180

                ele = 180 - ele

            elif ele < -90:

                ele += 180

                ele = -180 - ele



        # wrap azimuth value

        while np.abs(azi) > 180:

            azi = (azi + 180) % 360

            if azi < 0:

        while azi > 180:

            azi -= 360

        while azi <= -180:

            azi += 360

            azi -= 180



    # set -180 azimuth to 180

    if azi == -180:

        azi = 180



    return azi, ele
@@ -87,7 +84,6 @@ class EfapVertex: 
    Vertex data structure for EFAP

    Initialises a vertex from the given spherical coordinate pair, with a flag specifying if it is a ghost loudspeaker





    Parameters

    ----------

    azi : float
@@ -141,24 +137,21 @@ class EFAP: 


    Initialise EFAP data for computing panning gains





    Parameters

    ----------

    azimuths : np.ndarray

        Azimuth positions of the loudspeaker array

    elevations : npndarray

    elevations : np.ndarray

        Elevation postions of the loudspeaker array

    intensity_panning : bool

        Whether intensity panning is enabled or not



    Examples

    --------



    >>> from EFAP import EFAP

    >>> panner = EFAP([30, -30, 0, 110, -110], [0, 0, 0, 0, 0], False)

    >>> panner.pan(15, 45)

    array([0.66742381, 0.19069252, 0.66742381, 0.19069252, 0.19069252])



    """



    _EFAP_HULL_TOL = 1e-4  # tolerance for a point to be added to the convex hull
@@ -516,7 +509,6 @@ class EFAP: 
        """

        Compute panning gains for each vertex in the given polygon





        Parameters

        ----------

        azimuth : float
@@ -661,7 +653,6 @@ class EFAP: 
        C : float

            Third vertex of the triangle





        Returns

        -------

        bool
@@ -742,7 +733,8 @@ class EFAP: 
            return 0

        else:

            N = np.cross(P1 - P2, P1 - P3)

            return np.dot(X - P1, N / np.linalg.norm(N))

            eps = np.finfo(float).eps

            return np.dot(X - P1, N / (np.linalg.norm(N) + eps))



    def _flip_plane(self, surface: np.ndarray) -> np.ndarray:

        """
@@ -775,7 +767,6 @@ class EFAP: 
        """

        Compute gains for the requested panning position





        Parameters

        ----------

        azimuth : float
@@ -843,17 +834,19 @@ class EFAP: 
    """ public functions """



    def pan(

        self, azimuths: float, elevations: float, intensity_panning: bool = False

        self,

        azimuths: float,

        elevations: float,

        intensity_panning: Optional[bool] = False,

    ) -> np.ndarray:

        """

        Compute gains for the requested panning position





        Parameters

        ----------

        azimuth : float

        azimuths : float

            Azimuth of requested panning position

        elevation : float

        elevations : float

            Elevation of requested panning position

        intensity_panning : bool

            Flag whether to use intensity panning (Default is False == amplitude panning)
@@ -883,12 +876,10 @@ def main(args): 
    Parses a speaker layout text file and prints the panning gains

    for the requested position





    Parameters

    ----------

    args : tuple

    args : Namespace

        Command line arguments



    """



    speaker_positions = np.loadtxt(

scripts/pyaudio3dtools/__init__.py

+1 −1
Original line number Diff line number Diff line
@@ -48,5 +48,5 @@ from . import ( 
    spatialaudioconvert,

    spatialaudioformat,

    spatialmetadata,

    EFAP,

)
 
from .EFAP import EFAP

scripts/pyaudio3dtools/audio3dtools.py

+14 −13
Original line number Diff line number Diff line
@@ -35,8 +35,6 @@ import logging 
import os



from pyaudio3dtools import (

    audiofile,

    binauralrenderer,

    spatialaudioconvert,

    spatialaudioformat,

)
@@ -65,7 +63,7 @@ def main(): 
        "--outdir",

        required=True,

        type=str,

        help="output file *.wav or directory",

        help="output file *.wav, *.pcm or directory",

        default="out",

    )

    parser.add_argument(
@@ -134,8 +132,8 @@ def main(): 
        "-n",

        "--normalize",

        default=None,

        type=int,

        help="Normalize to given loudness with --LOUDNESS_TOOL (default = %(default)s)",

        type=float,

        help="Normalize to given loudness with BS 1770-4",

    )



    """ Miscellaneous or meta arguments """
@@ -165,11 +163,10 @@ def main(): 
        action="store_true",

    )

    parser.add_argument(

        "-lt",

        "--loudness_tool",

        default="bs1770demo",

        "-lf",

        "--loudness_format",

        type=str,

        help="Loudness tool to use: bs1770demo [default] or sv56demo (tool must be in $PATH or a path to the binary)",

        help="Loudness format: loudness tool renders to loudness format and scales such that this format has desired loudness (default output format)",

    )

    parser.add_argument(

        "-rn",
@@ -205,7 +202,9 @@ def main(): 
        if os.path.isdir(args.infiles):

            path = args.infiles

            audio_list = [

                os.path.join(path, f) for f in os.listdir(path) if f.endswith((".wav"))

                os.path.join(path, f)

                for f in os.listdir(path)

                if f.endswith(".wav") or f.endswith(".pcm") or f.endswith(".raw")

            ]

        else:

            audio_list = [args.infiles]
@@ -213,7 +212,9 @@ def main(): 
        outdir = args.outdir

        _, output_ext = os.path.splitext(os.path.basename(outdir))

        if (len(audio_list) == 1) and (

            (output_ext.lower() == ".wav") or (output_ext.lower() == ".pcm")

            (output_ext.lower() == ".wav")

            or (output_ext.lower() == ".pcm")

            or (output_ext.lower() == ".raw")

        ):

            outfile = outdir

        else:
@@ -246,7 +247,7 @@ def main(): 
                out_fs=args.outfs,

                out_fc=args.outfc,

                output_loudness=args.normalize,

                loudness_tool=args.loudness_tool,

                loudness_format=args.loudness_format,

                trajectory=args.trajectory,

                binaural_dataset=args.binaural_dataset,

            )
@@ -271,7 +272,7 @@ def main(): 
                    in_meta_files=args.metadata,

                    out_format="BINAURAL",

                    output_loudness=args.normalize,

                    loudness_tool=args.loudness_tool,

                    loudness_format=args.loudness_format,

                    trajectory=args.trajectory,

                    binaural_dataset=args.binaural_dataset,

                )