Skip to main content

An image source-based method used to simulate room transfer functions for arbitrary room shapes.

Project description

Diffraction Enhanced Image Source Method - Arbitrary Room Geometry (DEISM-ARG)

DOI

The code in this folder is able to solve the following problem:

A source and a receiver transducer with arbitrary directivity are mounted on one/two speakers; The local scattering and diffraction effects around the transducers result in complex directivity patterns. The directivity patterns can be obtained by analytical expressions, numerical simulations or measurements.

In DEISM-ARG, we can model the room transfer function between transducers mounted on one/two speakers using the image source method while incorporating the local diffraction effects around the transducers. The local diffraction effects are captured using spherical-harmonic directivity coefficients obtained on a sphere around the transducers. In addition to DEISM in shoebox rooms, DEISM-ARG can model more complex room shapes. However, for version 2.0, we now only supports convex shapes. In short, DEISM-ARG has the following features:

  1. Arbitrary directivities of the source and receiver
  2. Angle-dependent reflection coefficients, frequency- and wall-dependent impedance definition.
  3. Convex room shapes

image-20240812131054348

Preparation and installing

Build locally

  • In you encounter errors like "unrecognized arguments: deism_envs_exact.yml", please type the following commands manuelly in the command line.
  • Clone or download the repository to your local directory
  • Use the command conda env create -f deism_env.yml to create a Conda environment for the DEISM algorithms. If this does not work, try conda env create -f deism_env_exact.yml as the file deism_envs_exact.yml records the versions of all packages.
  • Activate the created environment by "conda activate DEISM"
  • Running pip install -e . will build the deism package including the c++ extensions locally. You can also modify the sources codes and check out the effects. In case you receiver errors like "ModuleNotFoundError: No module named 'pybind11" even after activated the conda environment, you can use python -m pip install -e . to try install again.
  • Run scripts in the examples folder.

Using pip to install remotely

  • Run pip install deism to install deism.

Additional heads up

  • You would need to an external LaTeX installation to render text properly when running the examples, since we enable plt.rcParams["text.usetex"] = True when plotting the simulated room transfer functions. Please refer to this page for more information Matplotlib usetex Documentation.

Running codes

Single set of parameters

The default parameters are defined in file configSingleParam.yml or configSingleParam_ARG.yml, depending on whether you want to run for shoebox rooms or more complicated room shapes. Take DEISM-ARG (configSingleParam_ARG.yml) as an example, there are two ways of running the codes:

  1. You can directly run deism_arg_singleparam_example.py in an IDE, which utilizes the parameters defined in configSingleParam_arg.yml.
  2. You can run deism_arg_singleparam_example.py from the command line after activating the conda environment. In addition, you can access help information quickly by python deism_arg_singleparam_example.py --help. You can then change the parameters based on the instructions from the help message, e.g., python deism_arg_singleparam_example.py -c 350 -zs 20 will change the parameter sound speed and the wall impedance. The new input value of the parameters then overrides the ones in file configSingleParam_arg.yml. After choosing the needed values, you can run the codes using, e.g., python deism_arg_singleparam_example.py -c 350 -zs 20 --run.
  3. You need to specify additionally the following parameters in the function init_parameters of deism_arg_singleparam_example.py:
    1. The vertices of the room
    2. If rotate the room w.r.t the origin. This can be useful if you want to have some comparisons with the rooms created using pyroomacoustics.
    3. The rotation angles of the room if it needs to be rotated.
  4. You can suppress all output information in the command line by adding flag "--quiet" or by setting the first parameters in the configuration.yml file SilentMode to 1.

Examples

DEISM-ARG

  • An example of running DEISM-ARG is examples/deism_arg_single_example.py
    • You can run this from IDEs or via the command line, as introduced in the previous section
  • An example of comparing different versions of DEISM-ARG is given in examples/deism_args_compare.py. A frequency- and wall-dependent impedance definition can also be applied for a convex room. The room transfer functions are compared among:
    • Original version (most computation-costly)
    • LC version (fastest)
    • Mix version (Trade-offs between Original and LC versions): Early reflections up to some changeable order (default is 2) are calculated using the original version and the higher orders are calculated using the LC version.
  • An example of comparing DEISM-ARG and pyroomacoustics is examples/deism_arg_pra_compare.py, Comparisons are done regarding if the following results are identical or mismatched only by a small deviation:
    • number of images
    • the positions of the images

DEISM for shoebox

We provide two examples for running the original DEISM for shoebox rooms

  • An example of running DEISM-shoebox is examples/deism_singleparam_example.py
    • You can either run this from IDEs or via the cmd
  • An example of comparing different DEISM versions are shown in examples/deisms_lc_mix_test.py. In this script, the following methods are compared
    • DEISM - original
    • DEISM - MIX (original + LC vectorized)
    • DEISM - LC vectorized
    • FEM as groundtruth (only for this specific parameter settings)

Tips

The example code only provides essential functionalities based on DEISM. For more complex scenarios, please contact the authors (zeyu.xu@audiolabs-erlangen.de) for support. In the following, a few important tips might be helpful for you:

  • If you want to simulate the scenario where both the source and receiver are positioned on the same speaker, you need to run DEISM for all reflection path except for the direct path.
  • It is recommended to set the distance at least 1m between the transducers and the walls.

Directivities

Modeling the directivities of the source and receiver in the room acoustics simulation is receiving increasing attention. The directivities of the source or receiver can include both the transducer directional properties and the local diffraction and scatterring effects caused by the enclosure where the transducers are mounted. Modern smart speakers are typical embodiments of such scenarios. Human heads are also a very common case.

Simple directivities

  • Monopole

Arbitrary directivities

Some key information should be provided if you want to include your own directivity data:

  1. Frequencies at which the directivities are simulated or measured. A 1D array.
  2. The spherical sampling directions around the transducer: azimuth from $0$ ( $+x$ direction) to $2 \pi$, inclination angle from $0$ ($+z$ direction) to $\pi$. A 2D array with size (number of directions, 2).
  3. The sampled pressure field at the specified directions and frequencies. A 2D array with size (number of frequencies, number of directions).
  4. The radius of the sampling sphere. A 1D array or float number.

For more information about directivity definition used in DEISM and DEISM-ARG, please refer to the following publication:

Zeyu Xu, Adrian Herzog, Alexander Lodermeyer, Emanuël A. P. Habets, Albert G. Prinn; Acoustic reciprocity in the spherical harmonic domain: A formulation for directional sources and receivers. JASA Express Lett. 1 December 2022; 2 (12): 124801. https://doi.org/10.1121/10.0016542

Contributors

  • M. Sc. Zeyu Xu
  • Songjiang Tan
  • M. Sc. Hasan Nazım Biçer
  • Dr. Albert Prinn
  • Prof. Dr. ir. Emanuël Habets

Academic publications

If you use this package in your research, please cite our paper:

Zeyu Xu, Adrian Herzog, Alexander Lodermeyer, Emanuël A. P. Habets, Albert G. Prinn; Simulating room transfer functions between transducers mounted on audio devices using a modified image source method. J. Acoust. Soc. Am. 1 January 2024; 155 (1): 343–357. https://doi.org/10.1121/10.0023935

Z. Xu, E.A.P. Habets and A.G. Prinn; Simulating sound fields in rooms with arbitrary geometries using the diffraction-enhanced image source method, Proc. of International Workshop on Acoustic Signal Enhancement (IWAENC), 2024.

Description of the codes and functions

configSingleParam_ARG.yml

In this file you define the default parameters for DEISM-ARG to run. Note that this file is different from the configSingleParam.yml on these parameters:

  • The dimensions are defined separately in the example script.
  • You also need to specify if you want to rotate the room, and the rotation angles as well.

configSingleParam.yml

In this file you define the default parameters for DEISM-shoebox to run.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

deism-2.1.1.tar.gz (2.4 MB view details)

Uploaded Source

Built Distributions

If you're not sure about the file name format, learn more about wheel file names.

deism-2.1.1-cp311-cp311-win_amd64.whl (3.5 MB view details)

Uploaded CPython 3.11Windows x86-64

deism-2.1.1-cp311-cp311-macosx_11_0_universal2.whl (3.6 MB view details)

Uploaded CPython 3.11macOS 11.0+ universal2 (ARM64, x86-64)

deism-2.1.1-cp311-cp311-macosx_10_9_universal2.whl (4.0 MB view details)

Uploaded CPython 3.11macOS 10.9+ universal2 (ARM64, x86-64)

deism-2.1.1-cp310-cp310-win_amd64.whl (3.5 MB view details)

Uploaded CPython 3.10Windows x86-64

deism-2.1.1-cp310-cp310-macosx_13_0_x86_64.whl (3.6 MB view details)

Uploaded CPython 3.10macOS 13.0+ x86-64

deism-2.1.1-cp310-cp310-macosx_11_0_universal2.whl (3.6 MB view details)

Uploaded CPython 3.10macOS 11.0+ universal2 (ARM64, x86-64)

deism-2.1.1-cp39-cp39-win_amd64.whl (3.5 MB view details)

Uploaded CPython 3.9Windows x86-64

deism-2.1.1-cp39-cp39-macosx_13_0_x86_64.whl (3.6 MB view details)

Uploaded CPython 3.9macOS 13.0+ x86-64

File details

Details for the file deism-2.1.1.tar.gz.

File metadata

  • Download URL: deism-2.1.1.tar.gz
  • Upload date:
  • Size: 2.4 MB
  • Tags: Source
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/6.1.0 CPython/3.12.9

File hashes

Hashes for deism-2.1.1.tar.gz
Algorithm Hash digest
SHA256 86d41982d7ca0c3508b3015a0496d08d3a89af26b6c5026e6033bd985ca47df3
MD5 bde4899e85b678245571721f06e39573
BLAKE2b-256 e36f009d39ee445fa3132258b00b5c8dec8ca1d0cefb48bea15b83474d327865

See more details on using hashes here.

Provenance

The following attestation bundles were made for deism-2.1.1.tar.gz:

Publisher: python-publish.yml on audiolabs/DEISM

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file deism-2.1.1-cp311-cp311-win_amd64.whl.

File metadata

  • Download URL: deism-2.1.1-cp311-cp311-win_amd64.whl
  • Upload date:
  • Size: 3.5 MB
  • Tags: CPython 3.11, Windows x86-64
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.1.0 CPython/3.11.9

File hashes

Hashes for deism-2.1.1-cp311-cp311-win_amd64.whl
Algorithm Hash digest
SHA256 e51366c7aa666c8847fdd7d02a138ee265c2a5bfefd7ccbc7fb7ee2a22a37292
MD5 374b4d0fc3de9f8cabefbc8f936947eb
BLAKE2b-256 30bfe0ff2fb82450ffdaa8cc40278a8cd5542132c5ea2c35a62200385bc31434

See more details on using hashes here.

File details

Details for the file deism-2.1.1-cp311-cp311-macosx_11_0_universal2.whl.

File metadata

File hashes

Hashes for deism-2.1.1-cp311-cp311-macosx_11_0_universal2.whl
Algorithm Hash digest
SHA256 be5bff81fdd5b0a32888284dc2d6b6946e969d16ea4d3ad646cfed7c4411e754
MD5 04edb126bbed58427e66e0de3e87f78b
BLAKE2b-256 2f0e3ca69cae693502b8c8514829afb5010056addfc39ade483eaa87de406d88

See more details on using hashes here.

File details

Details for the file deism-2.1.1-cp311-cp311-macosx_10_9_universal2.whl.

File metadata

File hashes

Hashes for deism-2.1.1-cp311-cp311-macosx_10_9_universal2.whl
Algorithm Hash digest
SHA256 7a318061a8ae2ccc594cd2c82dffbb340ce138dca4bd2c2832ae3c573bb9dca7
MD5 8a83ba16cec0ff194ce9f2ecf0db269c
BLAKE2b-256 8ebb55cc448d0749f6e2aa5b61a16af0dc04705d588d806ebb6be37200fc984f

See more details on using hashes here.

File details

Details for the file deism-2.1.1-cp310-cp310-win_amd64.whl.

File metadata

  • Download URL: deism-2.1.1-cp310-cp310-win_amd64.whl
  • Upload date:
  • Size: 3.5 MB
  • Tags: CPython 3.10, Windows x86-64
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.1.0 CPython/3.10.11

File hashes

Hashes for deism-2.1.1-cp310-cp310-win_amd64.whl
Algorithm Hash digest
SHA256 d1e227188600cce3efa27b7137c366d736f1d0634027acde37321aca6cf77be2
MD5 46954b5aa46c3ed62d2a1cfd714f1ade
BLAKE2b-256 f4b3a8517e558b74cf48c2460fd1610adfbfdee90315c6484180d744284af7f1

See more details on using hashes here.

File details

Details for the file deism-2.1.1-cp310-cp310-macosx_13_0_x86_64.whl.

File metadata

File hashes

Hashes for deism-2.1.1-cp310-cp310-macosx_13_0_x86_64.whl
Algorithm Hash digest
SHA256 bed4b4e13856f4461bf3973f895b7210f4e24281472780ecb634027ea7a07707
MD5 cc0acdc0457f974746c024eb5de9eacd
BLAKE2b-256 8c310510e12f18340679fa637537657cf3f108137ee453307451a4bf60c1b804

See more details on using hashes here.

File details

Details for the file deism-2.1.1-cp310-cp310-macosx_11_0_universal2.whl.

File metadata

File hashes

Hashes for deism-2.1.1-cp310-cp310-macosx_11_0_universal2.whl
Algorithm Hash digest
SHA256 faf55be1ed09193bccd6e192f8f7e7e3bc87a105aab0295fdaecf6dbfa01a075
MD5 bbcda3f0e926cfe2d7a0581ed4513a01
BLAKE2b-256 8e8d724ae07935b8aff5d61f0a3c746592a74511134ee8b404623b19dcaee1e1

See more details on using hashes here.

File details

Details for the file deism-2.1.1-cp39-cp39-win_amd64.whl.

File metadata

  • Download URL: deism-2.1.1-cp39-cp39-win_amd64.whl
  • Upload date:
  • Size: 3.5 MB
  • Tags: CPython 3.9, Windows x86-64
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.1.0 CPython/3.9.13

File hashes

Hashes for deism-2.1.1-cp39-cp39-win_amd64.whl
Algorithm Hash digest
SHA256 bdaf36a7032e299883074e1aac5442c857f298bddc132443504c89c46f533bcc
MD5 f08244a51e9f49534f31ebc9ece080a7
BLAKE2b-256 44ade01eb917adecfc6931cdfdf9a6d9a935c060c61aae2f293f6403e8dcc7bf

See more details on using hashes here.

File details

Details for the file deism-2.1.1-cp39-cp39-macosx_13_0_x86_64.whl.

File metadata

File hashes

Hashes for deism-2.1.1-cp39-cp39-macosx_13_0_x86_64.whl
Algorithm Hash digest
SHA256 9c9b519ec84a06f3ca23acd7a8105980fce58bfd74c685eb0adb73a57e46578a
MD5 926717bb1de726e10adba5ee2afe507e
BLAKE2b-256 43752d499c005736e119e0e0bfe274517b84ab5b92574eb9947f27395cd97d2f

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page